maximuspowers/muat-mean-std-classifier
Updated
Datasets:
example_id
int64 0
10k
| metadata
stringlengths 679
723
| classification_prompt
stringlengths 4.23k
10.2k
| classification_completion
stringclasses 14
values | classification_text
stringlengths 4.24k
10.2k
| improved_signature
stringlengths 2.11k
4.25k
| improved_model_weights
stringlengths 1.76k
5.04k
| training_metrics
stringlengths 1.46k
2.92k
|
|---|---|---|---|---|---|---|---|
0
|
{"target_pattern": "palindrome", "degraded_accuracy": 0.48, "improved_accuracy": 0.98, "improvement": 0.5, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 6, "neurons_per_layer": 7, "activation_type": "relu", "dropout_rate": 0.0, "random_seed": 2679, "learning_rate": 0.03008896643339405, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "palindrome", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["palindrome"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 6
Neurons per Layer: 7
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.453386,
-0.327474,
-0.157223,
0.456369,
0.643735
],
[
0.12622,
0.429354,
-0.240399,
-0.393445,
-0.106389
],
[
-0.11147,
-0.549566,
-0.162662,
-0.156643,
-0.175201
],
[
-0.363142,
-0.069774,
0.116957,
-0.013997,
0.706692
],
[
0.377102,
0.305662,
-0.360224,
-0.258711,
0.050089
],
[
-0.161948,
-0.358909,
-0.015153,
0.437877,
0.445758
],
[
-0.109331,
0.033178,
0.324673,
0.193858,
-0.184097
]
],
"network.0.bias": [
-0.119286,
0.111422,
-0.252262,
-0.234887,
0.168684,
-0.219954,
-0.279471
],
"network.2.weight": [
[
-0.285712,
0.218869,
0.205375,
0.076367,
-0.227966,
0.170337,
-0.164336
],
[
-0.062748,
-0.307176,
0.154941,
0.005569,
-0.186357,
-0.07184,
-0.157229
],
[
0.522343,
0.915817,
-0.046493,
0.489848,
0.243372,
0.375761,
0.044614
],
[
0.671219,
0.645983,
-0.024659,
0.565528,
0.512602,
0.790214,
-0.029846
],
[
0.319858,
-0.007872,
0.052752,
-0.024128,
-0.173129,
-0.322801,
-0.025422
],
[
0.805854,
0.588786,
-0.090574,
0.477788,
0.234788,
0.278425,
0.275044
],
[
-0.332876,
-0.43385,
0.147804,
-0.071303,
-0.258455,
0.059788,
-0.009553
]
],
"network.2.bias": [
-0.392752,
-0.295732,
-0.104868,
0.14474,
0.281554,
0.159095,
-0.174762
],
"network.4.weight": [
[
-0.176771,
-0.333024,
0.72132,
0.529057,
-0.276342,
0.033306,
-0.130475
],
[
0.133662,
-0.023771,
-0.371075,
-0.144869,
-0.441829,
-0.535646,
0.239691
],
[
-0.12163,
0.082703,
0.340026,
0.6903,
0.213804,
0.487876,
0.013686
],
[
-0.17301,
0.314767,
0.473747,
0.499127,
-0.179499,
0.583916,
0.133916
],
[
0.245098,
-0.042917,
0.649938,
0.641373,
0.14161,
0.804551,
-0.190475
],
[
-0.115826,
0.179191,
-0.505833,
-0.006435,
0.52513,
0.347838,
-0.443804
],
[
0.214556,
-0.099233,
0.543852,
0.742082,
-0.118269,
0.436566,
0.286135
]
],
"network.4.bias": [
-0.125316,
-0.00294,
0.225475,
-0.146175,
0.316137,
0.504589,
-0.210445
],
"network.6.weight": [
[
0.755015,
0.042379,
0.347016,
0.36354,
0.399971,
-0.512195,
0.601143
],
[
0.592505,
-0.343272,
0.045308,
0.494628,
0.268346,
-0.252532,
0.414568
],
[
0.490921,
-0.438342,
0.547926,
0.670926,
0.753076,
-0.481932,
0.683054
],
[
0.208314,
0.19906,
0.072818,
-0.248866,
-0.25011,
0.247131,
-0.218774
],
[
-0.243133,
0.162341,
-0.238087,
-0.144334,
-0.205474,
0.03245,
-0.303879
],
[
-0.394051,
0.140899,
0.191306,
0.088856,
0.310395,
0.342127,
-0.121957
],
[
-0.376836,
-0.030551,
-0.281656,
-0.307566,
-0.361284,
-0.109558,
0.096992
]
],
"network.6.bias": [
-0.012644,
-0.213544,
-0.130974,
-0.095416,
-0.13086,
0.729133,
-0.296097
],
"network.8.weight": [
[
-0.359781,
-0.009095,
0.122767,
-0.071826,
-0.042486,
0.199495,
0.034381
],
[
-0.312453,
-0.210575,
0.036223,
-0.326399,
-0.22013,
0.150529,
0.11431
],
[
0.224373,
-0.223163,
-0.015183,
0.019742,
0.011242,
0.790907,
-0.353785
],
[
-0.177242,
0.311048,
-0.316753,
-0.029666,
-0.343115,
0.01938,
-0.170107
],
[
-0.175515,
-0.025661,
0.180442,
-0.259101,
-0.209449,
0.497129,
-0.287058
],
[
-0.101743,
0.241576,
-0.424482,
-0.315128,
-0.005221,
-0.282459,
0.165427
],
[
0.386431,
0.3102,
0.760263,
-0.038546,
0.298923,
-0.616757,
0.024728
]
],
"network.8.bias": [
-0.363764,
-0.307377,
0.664011,
-0.243452,
0.525364,
0.03965,
0.129556
],
"network.10.weight": [
[
0.377958,
-0.348164,
0.113177,
0.209938,
-0.323088,
0.041111,
-0.149619
],
[
-0.072122,
-0.010595,
-0.398248,
0.259432,
0.090957,
-0.202259,
0.56982
],
[
-0.138257,
-0.17486,
0.320924,
0.0328,
-0.325217,
-0.25616,
-0.270992
],
[
0.282022,
-0.404823,
0.296393,
0.023418,
0.125639,
0.143354,
-0.262961
],
[
0.251087,
0.117399,
0.675095,
-0.372959,
0.520469,
0.145071,
-0.265112
],
[
0.303878,
0.421899,
-0.097009,
0.103215,
-0.56897,
0.093587,
0.510254
],
[
-0.097669,
0.530755,
-0.629361,
0.212001,
-0.12204,
0.008331,
0.03747
]
],
"network.10.bias": [
-0.306128,
0.133168,
-0.220066,
0.790848,
0.656986,
0.121512,
0.018336
],
"network.12.weight": [
[
-0.350767,
-0.653457,
-0.251655,
0.587816,
0.799047,
-0.589768,
0.133438
]
],
"network.12.bias": [
0.135511
]
}
## Activation Signature
### 0
mean: [1.292375, -0.426227, -2.284186, 0.271942, -0.050532, 0.381618, 0.511803]
std: [1.808385, 1.091531, 1.430994, 1.422668, 1.139070, 1.402958, 0.757621]
### 2
mean: [-0.779943, -0.685135, 1.601155, 2.425416, 0.404210, 2.259518, -0.881298]
std: [0.373547, 0.277124, 1.567902, 2.251955, 0.292369, 1.894588, 0.641896]
### 4
mean: [2.277668, -2.338019, 3.633575, 3.070626, 4.788693, 0.676361, 3.399807]
std: [2.332649, 1.998400, 3.041038, 2.930285, 3.998450, 0.225052, 3.319076]
### 6
mean: [7.700427, 5.345847, 10.643229, -1.894065, -3.989255, 2.100991, -4.598248]
std: [7.498685, 5.427755, 10.076451, 1.763134, 3.543395, 0.763215, 3.749759]
### 8
mean: [-1.457151, -3.138964, 2.697130, -3.273600, 2.001390, -4.561805, 11.561800]
std: [1.363895, 3.008224, 0.916671, 2.820644, 0.740204, 3.940965, 11.782944]
### 10
mean: [-2.380364, 5.840655, -3.143967, -1.203863, 0.448972, 4.630827, -1.489409]
std: [1.893654, 6.409157, 3.134876, 2.733996, 2.134031, 5.499527, 0.243057]
### 12
mean: [-5.419362]
std: [8.166819]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
palindrome
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 6
Neurons per Layer: 7
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.453386,
-0.327474,
-0.157223,
0.456369,
0.643735
],
[
0.12622,
0.429354,
-0.240399,
-0.393445,
-0.106389
],
[
-0.11147,
-0.549566,
-0.162662,
-0.156643,
-0.175201
],
[
-0.363142,
-0.069774,
0.116957,
-0.013997,
0.706692
],
[
0.377102,
0.305662,
-0.360224,
-0.258711,
0.050089
],
[
-0.161948,
-0.358909,
-0.015153,
0.437877,
0.445758
],
[
-0.109331,
0.033178,
0.324673,
0.193858,
-0.184097
]
],
"network.0.bias": [
-0.119286,
0.111422,
-0.252262,
-0.234887,
0.168684,
-0.219954,
-0.279471
],
"network.2.weight": [
[
-0.285712,
0.218869,
0.205375,
0.076367,
-0.227966,
0.170337,
-0.164336
],
[
-0.062748,
-0.307176,
0.154941,
0.005569,
-0.186357,
-0.07184,
-0.157229
],
[
0.522343,
0.915817,
-0.046493,
0.489848,
0.243372,
0.375761,
0.044614
],
[
0.671219,
0.645983,
-0.024659,
0.565528,
0.512602,
0.790214,
-0.029846
],
[
0.319858,
-0.007872,
0.052752,
-0.024128,
-0.173129,
-0.322801,
-0.025422
],
[
0.805854,
0.588786,
-0.090574,
0.477788,
0.234788,
0.278425,
0.275044
],
[
-0.332876,
-0.43385,
0.147804,
-0.071303,
-0.258455,
0.059788,
-0.009553
]
],
"network.2.bias": [
-0.392752,
-0.295732,
-0.104868,
0.14474,
0.281554,
0.159095,
-0.174762
],
"network.4.weight": [
[
-0.176771,
-0.333024,
0.72132,
0.529057,
-0.276342,
0.033306,
-0.130475
],
[
0.133662,
-0.023771,
-0.371075,
-0.144869,
-0.441829,
-0.535646,
0.239691
],
[
-0.12163,
0.082703,
0.340026,
0.6903,
0.213804,
0.487876,
0.013686
],
[
-0.17301,
0.314767,
0.473747,
0.499127,
-0.179499,
0.583916,
0.133916
],
[
0.245098,
-0.042917,
0.649938,
0.641373,
0.14161,
0.804551,
-0.190475
],
[
-0.115826,
0.179191,
-0.505833,
-0.006435,
0.52513,
0.347838,
-0.443804
],
[
0.214556,
-0.099233,
0.543852,
0.742082,
-0.118269,
0.436566,
0.286135
]
],
"network.4.bias": [
-0.125316,
-0.00294,
0.225475,
-0.146175,
0.316137,
0.504589,
-0.210445
],
"network.6.weight": [
[
0.755015,
0.042379,
0.347016,
0.36354,
0.399971,
-0.512195,
0.601143
],
[
0.592505,
-0.343272,
0.045308,
0.494628,
0.268346,
-0.252532,
0.414568
],
[
0.490921,
-0.438342,
0.547926,
0.670926,
0.753076,
-0.481932,
0.683054
],
[
0.208314,
0.19906,
0.072818,
-0.248866,
-0.25011,
0.247131,
-0.218774
],
[
-0.243133,
0.162341,
-0.238087,
-0.144334,
-0.205474,
0.03245,
-0.303879
],
[
-0.394051,
0.140899,
0.191306,
0.088856,
0.310395,
0.342127,
-0.121957
],
[
-0.376836,
-0.030551,
-0.281656,
-0.307566,
-0.361284,
-0.109558,
0.096992
]
],
"network.6.bias": [
-0.012644,
-0.213544,
-0.130974,
-0.095416,
-0.13086,
0.729133,
-0.296097
],
"network.8.weight": [
[
-0.359781,
-0.009095,
0.122767,
-0.071826,
-0.042486,
0.199495,
0.034381
],
[
-0.312453,
-0.210575,
0.036223,
-0.326399,
-0.22013,
0.150529,
0.11431
],
[
0.224373,
-0.223163,
-0.015183,
0.019742,
0.011242,
0.790907,
-0.353785
],
[
-0.177242,
0.311048,
-0.316753,
-0.029666,
-0.343115,
0.01938,
-0.170107
],
[
-0.175515,
-0.025661,
0.180442,
-0.259101,
-0.209449,
0.497129,
-0.287058
],
[
-0.101743,
0.241576,
-0.424482,
-0.315128,
-0.005221,
-0.282459,
0.165427
],
[
0.386431,
0.3102,
0.760263,
-0.038546,
0.298923,
-0.616757,
0.024728
]
],
"network.8.bias": [
-0.363764,
-0.307377,
0.664011,
-0.243452,
0.525364,
0.03965,
0.129556
],
"network.10.weight": [
[
0.377958,
-0.348164,
0.113177,
0.209938,
-0.323088,
0.041111,
-0.149619
],
[
-0.072122,
-0.010595,
-0.398248,
0.259432,
0.090957,
-0.202259,
0.56982
],
[
-0.138257,
-0.17486,
0.320924,
0.0328,
-0.325217,
-0.25616,
-0.270992
],
[
0.282022,
-0.404823,
0.296393,
0.023418,
0.125639,
0.143354,
-0.262961
],
[
0.251087,
0.117399,
0.675095,
-0.372959,
0.520469,
0.145071,
-0.265112
],
[
0.303878,
0.421899,
-0.097009,
0.103215,
-0.56897,
0.093587,
0.510254
],
[
-0.097669,
0.530755,
-0.629361,
0.212001,
-0.12204,
0.008331,
0.03747
]
],
"network.10.bias": [
-0.306128,
0.133168,
-0.220066,
0.790848,
0.656986,
0.121512,
0.018336
],
"network.12.weight": [
[
-0.350767,
-0.653457,
-0.251655,
0.587816,
0.799047,
-0.589768,
0.133438
]
],
"network.12.bias": [
0.135511
]
}
## Activation Signature
### 0
mean: [1.292375, -0.426227, -2.284186, 0.271942, -0.050532, 0.381618, 0.511803]
std: [1.808385, 1.091531, 1.430994, 1.422668, 1.139070, 1.402958, 0.757621]
### 2
mean: [-0.779943, -0.685135, 1.601155, 2.425416, 0.404210, 2.259518, -0.881298]
std: [0.373547, 0.277124, 1.567902, 2.251955, 0.292369, 1.894588, 0.641896]
### 4
mean: [2.277668, -2.338019, 3.633575, 3.070626, 4.788693, 0.676361, 3.399807]
std: [2.332649, 1.998400, 3.041038, 2.930285, 3.998450, 0.225052, 3.319076]
### 6
mean: [7.700427, 5.345847, 10.643229, -1.894065, -3.989255, 2.100991, -4.598248]
std: [7.498685, 5.427755, 10.076451, 1.763134, 3.543395, 0.763215, 3.749759]
### 8
mean: [-1.457151, -3.138964, 2.697130, -3.273600, 2.001390, -4.561805, 11.561800]
std: [1.363895, 3.008224, 0.916671, 2.820644, 0.740204, 3.940965, 11.782944]
### 10
mean: [-2.380364, 5.840655, -3.143967, -1.203863, 0.448972, 4.630827, -1.489409]
std: [1.893654, 6.409157, 3.134876, 2.733996, 2.134031, 5.499527, 0.243057]
### 12
mean: [-5.419362]
std: [8.166819]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
palindrome
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": 1.292374610900879, "std": 1.8083847761154175}, "1": {"mean": -0.4262273907661438, "std": 1.0915309190750122}, "2": {"mean": -2.284186363220215, "std": 1.4309940338134766}, "3": {"mean": 0.2719416618347168, "std": 1.42266845703125}, "4": {"mean": -0.05053213611245155, "std": 1.1390700340270996}, "5": {"mean": 0.3816179931163788, "std": 1.402957797050476}, "6": {"mean": 0.5118029713630676, "std": 0.7576205134391785}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": -0.7799431681632996, "std": 0.373546838760376}, "1": {"mean": -0.685134768486023, "std": 0.2771235704421997}, "2": {"mean": 1.6011545658111572, "std": 1.5679020881652832}, "3": {"mean": 2.4254157543182373, "std": 2.2519547939300537}, "4": {"mean": 0.40420985221862793, "std": 0.29236894845962524}, "5": {"mean": 2.2595183849334717, "std": 1.8945881128311157}, "6": {"mean": -0.8812984228134155, "std": 0.6418960690498352}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": 2.2776682376861572, "std": 2.332648515701294}, "1": {"mean": -2.3380191326141357, "std": 1.9984002113342285}, "2": {"mean": 3.6335747241973877, "std": 3.0410382747650146}, "3": {"mean": 3.0706264972686768, "std": 2.9302845001220703}, "4": {"mean": 4.788693428039551, "std": 3.9984498023986816}, "5": {"mean": 0.6763611435890198, "std": 0.22505198419094086}, "6": {"mean": 3.3998069763183594, "std": 3.3190762996673584}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": 7.700427055358887, "std": 7.498684883117676}, "1": {"mean": 5.345847129821777, "std": 5.427755355834961}, "2": {"mean": 10.643229484558105, "std": 10.076451301574707}, "3": {"mean": -1.894065499305725, "std": 1.7631340026855469}, "4": {"mean": -3.989255428314209, "std": 3.5433950424194336}, "5": {"mean": 2.1009905338287354, "std": 0.7632152438163757}, "6": {"mean": -4.59824800491333, "std": 3.7497591972351074}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -1.4571512937545776, "std": 1.3638947010040283}, "1": {"mean": -3.1389644145965576, "std": 3.0082240104675293}, "2": {"mean": 2.697129726409912, "std": 0.9166706204414368}, "3": {"mean": -3.273599863052368, "std": 2.8206441402435303}, "4": {"mean": 2.001389503479004, "std": 0.7402040362358093}, "5": {"mean": -4.561805248260498, "std": 3.940964937210083}, "6": {"mean": 11.561800003051758, "std": 11.782943725585938}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "10": {"neuron_profiles": {"0": {"mean": -2.380363702774048, "std": 1.8936537504196167}, "1": {"mean": 5.8406548500061035, "std": 6.4091572761535645}, "2": {"mean": -3.1439666748046875, "std": 3.134876012802124}, "3": {"mean": -1.2038630247116089, "std": 2.7339963912963867}, "4": {"mean": 0.44897210597991943, "std": 2.13403058052063}, "5": {"mean": 4.6308274269104, "std": 5.499527454376221}, "6": {"mean": -1.4894088506698608, "std": 0.24305716156959534}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "12": {"neuron_profiles": {"0": {"mean": -5.419361591339111, "std": 8.166818618774414}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 6, "neurons_per_layer": 7, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 6, "neurons_per_layer": 7, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[0.453386, -0.327474, -0.157223, 0.456369, 0.643735], [0.12622, 0.429354, -0.240399, -0.393445, -0.106389], [-0.11147, -0.549566, -0.162662, -0.156643, -0.175201], [-0.363142, -0.069774, 0.116957, -0.013997, 0.706692], [0.377102, 0.305662, -0.360224, -0.258711, 0.050089], [-0.161948, -0.358909, -0.015153, 0.437877, 0.445758], [-0.109331, 0.033178, 0.324673, 0.193858, -0.184097]], "network.0.bias": [-0.119286, 0.111422, -0.252262, -0.234887, 0.168684, -0.219954, -0.279471], "network.2.weight": [[-0.285712, 0.218869, 0.205375, 0.076367, -0.227966, 0.170337, -0.164336], [-0.062748, -0.307176, 0.154941, 0.005569, -0.186357, -0.07184, -0.157229], [0.522343, 0.915817, -0.046493, 0.489848, 0.243372, 0.375761, 0.044614], [0.671219, 0.645983, -0.024659, 0.565528, 0.512602, 0.790214, -0.029846], [0.319858, -0.007872, 0.052752, -0.024128, -0.173129, -0.322801, -0.025422], [0.805854, 0.588786, -0.090574, 0.477788, 0.234788, 0.278425, 0.275044], [-0.332876, -0.43385, 0.147804, -0.071303, -0.258455, 0.059788, -0.009553]], "network.2.bias": [-0.392752, -0.295732, -0.104868, 0.14474, 0.281554, 0.159095, -0.174762], "network.4.weight": [[-0.176771, -0.333024, 0.72132, 0.529057, -0.276342, 0.033306, -0.130475], [0.133662, -0.023771, -0.371075, -0.144869, -0.441829, -0.535646, 0.239691], [-0.12163, 0.082703, 0.340026, 0.6903, 0.213804, 0.487876, 0.013686], [-0.17301, 0.314767, 0.473747, 0.499127, -0.179499, 0.583916, 0.133916], [0.245098, -0.042917, 0.649938, 0.641373, 0.14161, 0.804551, -0.190475], [-0.115826, 0.179191, -0.505833, -0.006435, 0.52513, 0.347838, -0.443804], [0.214556, -0.099233, 0.543852, 0.742082, -0.118269, 0.436566, 0.286135]], "network.4.bias": [-0.125316, -0.00294, 0.225475, -0.146175, 0.316137, 0.504589, -0.210445], "network.6.weight": [[0.755015, 0.042379, 0.347016, 0.36354, 0.399971, -0.512195, 0.601143], [0.592505, -0.343272, 0.045308, 0.494628, 0.268346, -0.252532, 0.414568], [0.490921, -0.438342, 0.547926, 0.670926, 0.753076, -0.481932, 0.683054], [0.208314, 0.19906, 0.072818, -0.248866, -0.25011, 0.247131, -0.218774], [-0.243133, 0.162341, -0.238087, -0.144334, -0.205474, 0.03245, -0.303879], [-0.394051, 0.140899, 0.191306, 0.088856, 0.310395, 0.342127, -0.121957], [-0.376836, -0.030551, -0.281656, -0.307566, -0.361284, -0.109558, 0.096992]], "network.6.bias": [-0.012644, -0.213544, -0.130974, -0.095416, -0.13086, 0.729133, -0.296097], "network.8.weight": [[-0.359781, -0.009095, 0.122767, -0.071826, -0.042486, 0.199495, 0.034381], [-0.312453, -0.210575, 0.036223, -0.326399, -0.22013, 0.150529, 0.11431], [0.224373, -0.223163, -0.015183, 0.019742, 0.011242, 0.790907, -0.353785], [-0.177242, 0.311048, -0.316753, -0.029666, -0.343115, 0.01938, -0.170107], [-0.175515, -0.025661, 0.180442, -0.259101, -0.209449, 0.497129, -0.287058], [-0.101743, 0.241576, -0.424482, -0.315128, -0.005221, -0.282459, 0.165427], [0.386431, 0.3102, 0.760263, -0.038546, 0.298923, -0.616757, 0.024728]], "network.8.bias": [-0.363764, -0.307377, 0.664011, -0.243452, 0.525364, 0.03965, 0.129556], "network.10.weight": [[0.377958, -0.348164, 0.113177, 0.209938, -0.323088, 0.041111, -0.149619], [-0.072122, -0.010595, -0.398248, 0.259432, 0.090957, -0.202259, 0.56982], [-0.138257, -0.17486, 0.320924, 0.0328, -0.325217, -0.25616, -0.270992], [0.282022, -0.404823, 0.296393, 0.023418, 0.125639, 0.143354, -0.262961], [0.251087, 0.117399, 0.675095, -0.372959, 0.520469, 0.145071, -0.265112], [0.303878, 0.421899, -0.097009, 0.103215, -0.56897, 0.093587, 0.510254], [-0.097669, 0.530755, -0.629361, 0.212001, -0.12204, 0.008331, 0.03747]], "network.10.bias": [-0.306128, 0.133168, -0.220066, 0.790848, 0.656986, 0.121512, 0.018336], "network.12.weight": [[-0.350767, -0.653457, -0.251655, 0.587816, 0.799047, -0.589768, 0.133438]], "network.12.bias": [0.135511]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.6912154257297516, "train_acc": 0.58, "val_loss": 0.7252156138420105, "val_acc": 0.48}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.6803132891654968, "train_acc": 0.58, "val_loss": 0.7100004553794861, "val_acc": 0.48}, {"stage": "degraded", "epoch": 2, "global_epoch": 2, "train_loss": 0.6794183850288391, "train_acc": 0.58, "val_loss": 0.6856794953346252, "val_acc": 0.48}, {"stage": "improved", "epoch": 0, "global_epoch": 3, "train_loss": 0.6625008881092072, "train_acc": 0.505, "val_loss": 0.5861369967460632, "val_acc": 0.48}, {"stage": "improved", "epoch": 1, "global_epoch": 4, "train_loss": 0.5708945691585541, "train_acc": 0.61, "val_loss": 0.45573386549949646, "val_acc": 0.9}, {"stage": "improved", "epoch": 2, "global_epoch": 5, "train_loss": 0.4603542387485504, "train_acc": 0.885, "val_loss": 0.38293492794036865, "val_acc": 0.88}, {"stage": "improved", "epoch": 3, "global_epoch": 6, "train_loss": 0.3767639994621277, "train_acc": 0.89, "val_loss": 0.3039224445819855, "val_acc": 0.92}, {"stage": "improved", "epoch": 4, "global_epoch": 7, "train_loss": 0.31747712194919586, "train_acc": 0.885, "val_loss": 0.2419426292181015, "val_acc": 0.96}, {"stage": "improved", "epoch": 5, "global_epoch": 8, "train_loss": 0.2613602429628372, "train_acc": 0.895, "val_loss": 0.22014766931533813, "val_acc": 0.94}, {"stage": "improved", "epoch": 6, "global_epoch": 9, "train_loss": 0.24601513147354126, "train_acc": 0.9, "val_loss": 0.16306976974010468, "val_acc": 0.92}, {"stage": "improved", "epoch": 7, "global_epoch": 10, "train_loss": 0.235054612159729, "train_acc": 0.905, "val_loss": 0.15598948299884796, "val_acc": 0.98}, {"stage": "improved", "epoch": 8, "global_epoch": 11, "train_loss": 0.22074957191944122, "train_acc": 0.905, "val_loss": 0.12956710159778595, "val_acc": 0.94}, {"stage": "improved", "epoch": 9, "global_epoch": 12, "train_loss": 0.21819238364696503, "train_acc": 0.92, "val_loss": 0.12354502081871033, "val_acc": 0.96}], "summary": {"total_epochs": 13, "degraded_epochs": 3, "improved_epochs": 10, "patterns": ["palindrome"], "degraded_stage": {"initial_val_loss": 0.7252156138420105, "final_val_loss": 0.6856794953346252, "initial_val_acc": 0.48, "final_val_acc": 0.48, "best_val_acc": 0.48}, "improved_stage": {"initial_val_loss": 0.5861369967460632, "final_val_loss": 0.12354502081871033, "initial_val_acc": 0.48, "final_val_acc": 0.96, "best_val_acc": 0.98, "best_epoch": 10}, "improvement": 0.5, "first_improvement_epoch": 2}}
|
1
|
{"target_pattern": "alternating", "degraded_accuracy": 0.52, "improved_accuracy": 0.92, "improvement": 0.4, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 7, "activation_type": "gelu", "dropout_rate": 0.0, "random_seed": 9451, "learning_rate": 0.07352170370310572, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "alternating", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["alternating"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 7
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.350749,
0.022741,
0.24561,
0.369145,
0.543818
],
[
-0.784593,
-0.505623,
0.047746,
-0.131884,
-0.051516
],
[
-0.023241,
-0.080901,
-0.369471,
0.099241,
0.402252
],
[
-0.223858,
0.376454,
-0.595125,
0.390772,
-0.327672
],
[
-0.154215,
0.640965,
0.136615,
0.515237,
0.278974
],
[
-0.233706,
-0.19494,
0.353523,
0.112396,
-0.411489
],
[
0.418931,
-0.412868,
0.336435,
-0.763107,
-0.069389
]
],
"network.0.bias": [
0.36142,
0.084081,
-0.340971,
0.395617,
0.166324,
-0.2408,
-0.482828
],
"network.2.weight": [
[
-0.236004,
0.578259,
-0.289967,
0.515971,
0.072462,
0.011791,
0.16048
],
[
0.474862,
-0.459772,
0.01287,
-0.344333,
0.015363,
0.03543,
-0.589452
],
[
-0.118009,
0.40436,
0.163597,
-0.1563,
-0.099906,
0.203053,
0.044845
],
[
0.042569,
-0.112881,
0.704737,
-0.75616,
0.606623,
0.660571,
-0.392786
],
[
0.322126,
-0.40847,
0.086208,
0.554093,
-0.200485,
-0.243866,
0.55913
],
[
0.483438,
-0.236135,
0.501725,
-0.380317,
0.332086,
0.238996,
-0.293392
],
[
0.043822,
0.22006,
-0.005235,
0.534647,
-0.091811,
-0.034812,
0.310352
]
],
"network.2.bias": [
-0.123138,
0.187383,
0.282959,
0.130568,
0.009213,
0.350078,
0.063918
],
"network.4.weight": [
[
-0.369474,
0.387959,
-0.210282,
-0.195506,
-0.254928,
0.577384,
-0.30998
],
[
-0.200764,
0.136217,
-0.102683,
0.601136,
-0.304961,
0.364274,
-0.43103
],
[
-0.126448,
0.40362,
0.015479,
0.696791,
-0.180078,
0.308092,
-0.099349
],
[
0.045698,
-0.635704,
-0.168412,
0.468472,
0.128407,
-0.398191,
-0.060673
],
[
-0.619535,
0.150121,
-0.30259,
0.213474,
0.683084,
-0.163303,
-0.053653
],
[
-0.564362,
0.135669,
0.107934,
-0.10099,
0.539508,
-0.177831,
0.157239
],
[
-0.153653,
-0.436977,
-0.222605,
-0.152053,
-0.144703,
-0.516253,
0.453713
]
],
"network.4.bias": [
-0.260694,
0.235258,
0.245827,
-0.222018,
-0.114456,
-0.146646,
-0.206067
],
"network.6.weight": [
[
-0.258117,
-0.174745,
0.038918,
-0.545319,
0.07179,
-0.25871,
-0.161076
],
[
0.271989,
0.407153,
0.067816,
0.240921,
-0.023037,
0.32908,
0.291896
],
[
-0.11323,
0.108879,
-0.211443,
-0.174271,
0.124389,
0.146353,
0.11751
],
[
-0.117422,
0.604238,
0.072596,
0.29033,
0.313296,
-0.144542,
0.535626
],
[
-0.347906,
-0.428626,
0.284223,
-0.358097,
-0.211131,
-0.422963,
-0.009197
],
[
0.124001,
0.004126,
-0.295901,
-0.228066,
-0.455679,
-0.345138,
-0.099782
],
[
0.504156,
0.310465,
0.521353,
0.133778,
-0.327467,
-0.366576,
0.374892
]
],
"network.6.bias": [
-0.076262,
-0.219414,
0.261658,
-0.005811,
0.06013,
-0.102406,
0.251035
],
"network.8.weight": [
[
0.160282,
0.468223,
-0.061712,
0.175455,
0.10077,
0.006619,
-0.026136
],
[
-0.275662,
0.069069,
-0.334605,
0.056033,
-0.12303,
-0.346975,
0.36876
],
[
-0.015352,
-0.441747,
-0.142609,
-0.394476,
0.565378,
-0.039106,
-0.018703
],
[
-0.152891,
0.458172,
-0.195141,
0.598307,
-0.314712,
-0.038004,
0.629842
],
[
0.589618,
-0.076678,
-0.292796,
-0.156061,
0.004389,
-0.19163,
-0.310142
],
[
-0.009763,
-0.124552,
-0.277456,
-0.088162,
0.344165,
-0.064033,
-0.13644
],
[
-0.288017,
-0.268217,
0.356369,
-0.022197,
-0.255733,
-0.351994,
-0.13463
]
],
"network.8.bias": [
0.243465,
0.324187,
-0.014752,
-0.059377,
0.175747,
0.255096,
0.215125
],
"network.10.weight": [
[
-0.014447,
-0.375855,
0.119083,
-0.492942,
0.129516,
0.301704,
0.201898
]
],
"network.10.bias": [
0.264722
]
}
## Activation Signature
### 0
mean: [1.941744, -2.060277, -0.580170, -0.006271, 2.876730, -0.414864, -1.733074]
std: [1.540038, 2.131134, 0.994359, 1.861390, 1.969188, 1.032517, 2.210172]
### 2
mean: [-0.098283, 0.884047, -0.298041, 1.566564, 0.460217, 2.025386, 0.224169]
std: [0.679813, 0.858557, 0.424365, 1.426804, 0.702743, 1.413017, 0.528944]
### 4
mean: [0.687249, 1.737890, 2.126264, -0.746205, 0.258241, -0.328324, -1.768268]
std: [1.027876, 1.629525, 1.784214, 0.525884, 0.387891, 0.408743, 1.333594]
### 6
mean: [-0.364685, 0.730973, -0.040689, 1.093562, -0.303306, -0.654163, 2.126337]
std: [0.427067, 1.010593, 0.301358, 1.103774, 0.510473, 0.513145, 1.895594]
### 8
mean: [0.690842, 1.261036, -0.801140, 2.231313, -0.716119, -0.219866, -0.203726]
std: [0.575193, 0.904849, 0.914128, 2.350589, 0.837312, 0.471279, 0.535711]
### 10
mean: [-1.292513]
std: [1.588143]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
alternating
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 7
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.350749,
0.022741,
0.24561,
0.369145,
0.543818
],
[
-0.784593,
-0.505623,
0.047746,
-0.131884,
-0.051516
],
[
-0.023241,
-0.080901,
-0.369471,
0.099241,
0.402252
],
[
-0.223858,
0.376454,
-0.595125,
0.390772,
-0.327672
],
[
-0.154215,
0.640965,
0.136615,
0.515237,
0.278974
],
[
-0.233706,
-0.19494,
0.353523,
0.112396,
-0.411489
],
[
0.418931,
-0.412868,
0.336435,
-0.763107,
-0.069389
]
],
"network.0.bias": [
0.36142,
0.084081,
-0.340971,
0.395617,
0.166324,
-0.2408,
-0.482828
],
"network.2.weight": [
[
-0.236004,
0.578259,
-0.289967,
0.515971,
0.072462,
0.011791,
0.16048
],
[
0.474862,
-0.459772,
0.01287,
-0.344333,
0.015363,
0.03543,
-0.589452
],
[
-0.118009,
0.40436,
0.163597,
-0.1563,
-0.099906,
0.203053,
0.044845
],
[
0.042569,
-0.112881,
0.704737,
-0.75616,
0.606623,
0.660571,
-0.392786
],
[
0.322126,
-0.40847,
0.086208,
0.554093,
-0.200485,
-0.243866,
0.55913
],
[
0.483438,
-0.236135,
0.501725,
-0.380317,
0.332086,
0.238996,
-0.293392
],
[
0.043822,
0.22006,
-0.005235,
0.534647,
-0.091811,
-0.034812,
0.310352
]
],
"network.2.bias": [
-0.123138,
0.187383,
0.282959,
0.130568,
0.009213,
0.350078,
0.063918
],
"network.4.weight": [
[
-0.369474,
0.387959,
-0.210282,
-0.195506,
-0.254928,
0.577384,
-0.30998
],
[
-0.200764,
0.136217,
-0.102683,
0.601136,
-0.304961,
0.364274,
-0.43103
],
[
-0.126448,
0.40362,
0.015479,
0.696791,
-0.180078,
0.308092,
-0.099349
],
[
0.045698,
-0.635704,
-0.168412,
0.468472,
0.128407,
-0.398191,
-0.060673
],
[
-0.619535,
0.150121,
-0.30259,
0.213474,
0.683084,
-0.163303,
-0.053653
],
[
-0.564362,
0.135669,
0.107934,
-0.10099,
0.539508,
-0.177831,
0.157239
],
[
-0.153653,
-0.436977,
-0.222605,
-0.152053,
-0.144703,
-0.516253,
0.453713
]
],
"network.4.bias": [
-0.260694,
0.235258,
0.245827,
-0.222018,
-0.114456,
-0.146646,
-0.206067
],
"network.6.weight": [
[
-0.258117,
-0.174745,
0.038918,
-0.545319,
0.07179,
-0.25871,
-0.161076
],
[
0.271989,
0.407153,
0.067816,
0.240921,
-0.023037,
0.32908,
0.291896
],
[
-0.11323,
0.108879,
-0.211443,
-0.174271,
0.124389,
0.146353,
0.11751
],
[
-0.117422,
0.604238,
0.072596,
0.29033,
0.313296,
-0.144542,
0.535626
],
[
-0.347906,
-0.428626,
0.284223,
-0.358097,
-0.211131,
-0.422963,
-0.009197
],
[
0.124001,
0.004126,
-0.295901,
-0.228066,
-0.455679,
-0.345138,
-0.099782
],
[
0.504156,
0.310465,
0.521353,
0.133778,
-0.327467,
-0.366576,
0.374892
]
],
"network.6.bias": [
-0.076262,
-0.219414,
0.261658,
-0.005811,
0.06013,
-0.102406,
0.251035
],
"network.8.weight": [
[
0.160282,
0.468223,
-0.061712,
0.175455,
0.10077,
0.006619,
-0.026136
],
[
-0.275662,
0.069069,
-0.334605,
0.056033,
-0.12303,
-0.346975,
0.36876
],
[
-0.015352,
-0.441747,
-0.142609,
-0.394476,
0.565378,
-0.039106,
-0.018703
],
[
-0.152891,
0.458172,
-0.195141,
0.598307,
-0.314712,
-0.038004,
0.629842
],
[
0.589618,
-0.076678,
-0.292796,
-0.156061,
0.004389,
-0.19163,
-0.310142
],
[
-0.009763,
-0.124552,
-0.277456,
-0.088162,
0.344165,
-0.064033,
-0.13644
],
[
-0.288017,
-0.268217,
0.356369,
-0.022197,
-0.255733,
-0.351994,
-0.13463
]
],
"network.8.bias": [
0.243465,
0.324187,
-0.014752,
-0.059377,
0.175747,
0.255096,
0.215125
],
"network.10.weight": [
[
-0.014447,
-0.375855,
0.119083,
-0.492942,
0.129516,
0.301704,
0.201898
]
],
"network.10.bias": [
0.264722
]
}
## Activation Signature
### 0
mean: [1.941744, -2.060277, -0.580170, -0.006271, 2.876730, -0.414864, -1.733074]
std: [1.540038, 2.131134, 0.994359, 1.861390, 1.969188, 1.032517, 2.210172]
### 2
mean: [-0.098283, 0.884047, -0.298041, 1.566564, 0.460217, 2.025386, 0.224169]
std: [0.679813, 0.858557, 0.424365, 1.426804, 0.702743, 1.413017, 0.528944]
### 4
mean: [0.687249, 1.737890, 2.126264, -0.746205, 0.258241, -0.328324, -1.768268]
std: [1.027876, 1.629525, 1.784214, 0.525884, 0.387891, 0.408743, 1.333594]
### 6
mean: [-0.364685, 0.730973, -0.040689, 1.093562, -0.303306, -0.654163, 2.126337]
std: [0.427067, 1.010593, 0.301358, 1.103774, 0.510473, 0.513145, 1.895594]
### 8
mean: [0.690842, 1.261036, -0.801140, 2.231313, -0.716119, -0.219866, -0.203726]
std: [0.575193, 0.904849, 0.914128, 2.350589, 0.837312, 0.471279, 0.535711]
### 10
mean: [-1.292513]
std: [1.588143]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
alternating
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": 1.9417439699172974, "std": 1.5400378704071045}, "1": {"mean": -2.060276746749878, "std": 2.131134033203125}, "2": {"mean": -0.5801698565483093, "std": 0.9943591952323914}, "3": {"mean": -0.0062707290053367615, "std": 1.861390233039856}, "4": {"mean": 2.876730442047119, "std": 1.969187617301941}, "5": {"mean": -0.4148639142513275, "std": 1.0325170755386353}, "6": {"mean": -1.7330735921859741, "std": 2.210172176361084}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": -0.09828302264213562, "std": 0.6798130869865417}, "1": {"mean": 0.8840467929840088, "std": 0.8585572838783264}, "2": {"mean": -0.2980407774448395, "std": 0.424365371465683}, "3": {"mean": 1.5665640830993652, "std": 1.4268038272857666}, "4": {"mean": 0.4602169394493103, "std": 0.7027427554130554}, "5": {"mean": 2.025385618209839, "std": 1.4130165576934814}, "6": {"mean": 0.2241688221693039, "std": 0.5289435982704163}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": 0.6872493624687195, "std": 1.0278756618499756}, "1": {"mean": 1.7378902435302734, "std": 1.6295251846313477}, "2": {"mean": 2.1262643337249756, "std": 1.7842140197753906}, "3": {"mean": -0.7462049722671509, "std": 0.5258837938308716}, "4": {"mean": 0.2582411766052246, "std": 0.38789135217666626}, "5": {"mean": -0.32832419872283936, "std": 0.40874314308166504}, "6": {"mean": -1.768268346786499, "std": 1.3335939645767212}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": -0.36468467116355896, "std": 0.427067369222641}, "1": {"mean": 0.7309728860855103, "std": 1.0105925798416138}, "2": {"mean": -0.040688831359148026, "std": 0.3013584315776825}, "3": {"mean": 1.0935624837875366, "std": 1.103773593902588}, "4": {"mean": -0.3033055067062378, "std": 0.5104734897613525}, "5": {"mean": -0.6541626453399658, "std": 0.5131453275680542}, "6": {"mean": 2.1263365745544434, "std": 1.8955938816070557}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": 0.6908419728279114, "std": 0.575192928314209}, "1": {"mean": 1.2610357999801636, "std": 0.9048486351966858}, "2": {"mean": -0.801139771938324, "std": 0.9141275882720947}, "3": {"mean": 2.2313129901885986, "std": 2.3505892753601074}, "4": {"mean": -0.7161186337471008, "std": 0.8373115062713623}, "5": {"mean": -0.21986648440361023, "std": 0.47127893567085266}, "6": {"mean": -0.20372597873210907, "std": 0.5357106924057007}}, "layer_info": {"num_neurons": 7, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "10": {"neuron_profiles": {"0": {"mean": -1.292513132095337, "std": 1.5881426334381104}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 7, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 7, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[-0.350749, 0.022741, 0.24561, 0.369145, 0.543818], [-0.784593, -0.505623, 0.047746, -0.131884, -0.051516], [-0.023241, -0.080901, -0.369471, 0.099241, 0.402252], [-0.223858, 0.376454, -0.595125, 0.390772, -0.327672], [-0.154215, 0.640965, 0.136615, 0.515237, 0.278974], [-0.233706, -0.19494, 0.353523, 0.112396, -0.411489], [0.418931, -0.412868, 0.336435, -0.763107, -0.069389]], "network.0.bias": [0.36142, 0.084081, -0.340971, 0.395617, 0.166324, -0.2408, -0.482828], "network.2.weight": [[-0.236004, 0.578259, -0.289967, 0.515971, 0.072462, 0.011791, 0.16048], [0.474862, -0.459772, 0.01287, -0.344333, 0.015363, 0.03543, -0.589452], [-0.118009, 0.40436, 0.163597, -0.1563, -0.099906, 0.203053, 0.044845], [0.042569, -0.112881, 0.704737, -0.75616, 0.606623, 0.660571, -0.392786], [0.322126, -0.40847, 0.086208, 0.554093, -0.200485, -0.243866, 0.55913], [0.483438, -0.236135, 0.501725, -0.380317, 0.332086, 0.238996, -0.293392], [0.043822, 0.22006, -0.005235, 0.534647, -0.091811, -0.034812, 0.310352]], "network.2.bias": [-0.123138, 0.187383, 0.282959, 0.130568, 0.009213, 0.350078, 0.063918], "network.4.weight": [[-0.369474, 0.387959, -0.210282, -0.195506, -0.254928, 0.577384, -0.30998], [-0.200764, 0.136217, -0.102683, 0.601136, -0.304961, 0.364274, -0.43103], [-0.126448, 0.40362, 0.015479, 0.696791, -0.180078, 0.308092, -0.099349], [0.045698, -0.635704, -0.168412, 0.468472, 0.128407, -0.398191, -0.060673], [-0.619535, 0.150121, -0.30259, 0.213474, 0.683084, -0.163303, -0.053653], [-0.564362, 0.135669, 0.107934, -0.10099, 0.539508, -0.177831, 0.157239], [-0.153653, -0.436977, -0.222605, -0.152053, -0.144703, -0.516253, 0.453713]], "network.4.bias": [-0.260694, 0.235258, 0.245827, -0.222018, -0.114456, -0.146646, -0.206067], "network.6.weight": [[-0.258117, -0.174745, 0.038918, -0.545319, 0.07179, -0.25871, -0.161076], [0.271989, 0.407153, 0.067816, 0.240921, -0.023037, 0.32908, 0.291896], [-0.11323, 0.108879, -0.211443, -0.174271, 0.124389, 0.146353, 0.11751], [-0.117422, 0.604238, 0.072596, 0.29033, 0.313296, -0.144542, 0.535626], [-0.347906, -0.428626, 0.284223, -0.358097, -0.211131, -0.422963, -0.009197], [0.124001, 0.004126, -0.295901, -0.228066, -0.455679, -0.345138, -0.099782], [0.504156, 0.310465, 0.521353, 0.133778, -0.327467, -0.366576, 0.374892]], "network.6.bias": [-0.076262, -0.219414, 0.261658, -0.005811, 0.06013, -0.102406, 0.251035], "network.8.weight": [[0.160282, 0.468223, -0.061712, 0.175455, 0.10077, 0.006619, -0.026136], [-0.275662, 0.069069, -0.334605, 0.056033, -0.12303, -0.346975, 0.36876], [-0.015352, -0.441747, -0.142609, -0.394476, 0.565378, -0.039106, -0.018703], [-0.152891, 0.458172, -0.195141, 0.598307, -0.314712, -0.038004, 0.629842], [0.589618, -0.076678, -0.292796, -0.156061, 0.004389, -0.19163, -0.310142], [-0.009763, -0.124552, -0.277456, -0.088162, 0.344165, -0.064033, -0.13644], [-0.288017, -0.268217, 0.356369, -0.022197, -0.255733, -0.351994, -0.13463]], "network.8.bias": [0.243465, 0.324187, -0.014752, -0.059377, 0.175747, 0.255096, 0.215125], "network.10.weight": [[-0.014447, -0.375855, 0.119083, -0.492942, 0.129516, 0.301704, 0.201898]], "network.10.bias": [0.264722]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.6916628181934357, "train_acc": 0.545, "val_loss": 0.6948902606964111, "val_acc": 0.52}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.6612586081027985, "train_acc": 0.56, "val_loss": 0.6322479248046875, "val_acc": 0.52}, {"stage": "improved", "epoch": 0, "global_epoch": 2, "train_loss": 0.5553124845027924, "train_acc": 0.55, "val_loss": 0.39744308590888977, "val_acc": 0.92}, {"stage": "improved", "epoch": 1, "global_epoch": 3, "train_loss": 0.3223879337310791, "train_acc": 0.945, "val_loss": 0.4029606580734253, "val_acc": 0.82}, {"stage": "improved", "epoch": 2, "global_epoch": 4, "train_loss": 0.3463154584169388, "train_acc": 0.875, "val_loss": 0.4694277048110962, "val_acc": 0.8}, {"stage": "improved", "epoch": 3, "global_epoch": 5, "train_loss": 0.28306517004966736, "train_acc": 0.89, "val_loss": 0.49492138624191284, "val_acc": 0.82}], "summary": {"total_epochs": 6, "degraded_epochs": 2, "improved_epochs": 4, "patterns": ["alternating"], "degraded_stage": {"initial_val_loss": 0.6948902606964111, "final_val_loss": 0.6322479248046875, "initial_val_acc": 0.52, "final_val_acc": 0.52, "best_val_acc": 0.52}, "improved_stage": {"initial_val_loss": 0.39744308590888977, "final_val_loss": 0.49492138624191284, "initial_val_acc": 0.92, "final_val_acc": 0.82, "best_val_acc": 0.92, "best_epoch": 2}, "improvement": 0.4, "first_improvement_epoch": 1}}
|
2
|
{"target_pattern": "increasing_pairs", "degraded_accuracy": 0.5, "improved_accuracy": 0.9, "improvement": 0.4, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 6, "activation_type": "gelu", "dropout_rate": 0.0, "random_seed": 7902, "learning_rate": 0.019119242316001303, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "increasing_pairs", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["increasing_pairs"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 6
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.306744,
-0.107353,
0.323492,
-0.263617,
-0.071321
],
[
-0.136755,
0.031543,
-0.507803,
-0.48734,
-0.228265
],
[
0.462246,
0.20675,
0.034836,
-0.200921,
-0.218379
],
[
0.441719,
-0.246824,
0.40355,
-0.250275,
-0.173721
],
[
-0.27147,
0.000759,
0.24793,
0.206822,
-0.30087
],
[
-0.128065,
-0.089901,
0.207173,
0.481966,
0.683192
]
],
"network.0.bias": [
0.017415,
0.209661,
0.312007,
-0.179297,
-0.156691,
0.609966
],
"network.2.weight": [
[
0.146409,
0.229788,
-0.182541,
0.233177,
0.00381,
-0.302754
],
[
-0.376353,
-0.162428,
-0.01784,
0.144334,
0.164582,
-0.331682
],
[
-0.166265,
0.220298,
-0.286452,
-0.325655,
-0.423274,
0.209143
],
[
-0.047387,
-0.197965,
-0.35189,
0.059013,
-0.347397,
0.11409
],
[
-0.090832,
0.197845,
-0.117846,
0.284988,
0.020934,
-0.291669
],
[
0.241248,
0.013841,
-0.168016,
-0.224182,
0.337959,
-0.371723
]
],
"network.2.bias": [
-0.228908,
-0.139118,
-0.090114,
-0.123791,
-0.273926,
0.146292
],
"network.4.weight": [
[
-0.538209,
-0.300712,
-0.563901,
-0.023583,
-0.404277,
-0.498438
],
[
0.10558,
0.13363,
0.195469,
0.375287,
0.192972,
0.610911
],
[
-0.592715,
-0.484534,
-0.199585,
0.116719,
-0.29873,
-0.101867
],
[
-0.398807,
0.122273,
-0.381404,
-0.213997,
-0.182524,
-0.129994
],
[
-0.016247,
0.068468,
0.350695,
-0.191235,
0.138635,
0.538008
],
[
-0.366761,
-0.22621,
-0.200696,
-0.589808,
-0.541964,
-0.457728
]
],
"network.4.bias": [
0.559085,
-0.152053,
-0.076749,
0.445926,
-0.223831,
0.474309
],
"network.6.weight": [
[
0.327886,
-0.496222,
0.331148,
0.227621,
-0.5305,
0.64235
],
[
-0.433197,
0.39602,
-0.308265,
0.236144,
0.145873,
-0.373335
],
[
-0.364165,
0.225038,
-0.295511,
-0.019409,
0.111191,
0.04238
],
[
0.615553,
0.1129,
-0.122227,
0.185108,
0.160538,
0.614453
],
[
0.08013,
0.390056,
-0.054861,
0.214988,
0.340561,
-0.006452
],
[
0.319114,
-0.343264,
0.037997,
0.494555,
-0.653354,
0.188986
]
],
"network.6.bias": [
0.242986,
0.058184,
-0.020256,
0.224645,
-0.06052,
0.222313
],
"network.8.weight": [
[
-0.528034,
0.223621,
0.396886,
-0.253046,
0.005106,
-0.631389
]
],
"network.8.bias": [
0.263716
]
}
## Activation Signature
### 0
mean: [0.225986, -2.290389, 0.636561, 0.012643, 0.097207, 2.595692]
std: [1.142856, 1.645565, 1.221557, 1.430944, 0.902008, 1.787508]
### 2
mean: [-0.954601, -1.048196, -0.118633, -0.171197, -1.005633, -0.810383]
std: [0.578908, 0.621705, 0.813907, 0.446660, 0.530225, 0.682586]
### 4
mean: [0.697518, -0.252426, 0.084479, 0.466904, -0.259491, 0.649963]
std: [0.264418, 0.167469, 0.103890, 0.196127, 0.117419, 0.224882]
### 6
mean: [0.931000, -0.352502, -0.251233, 0.897562, -0.019905, 0.751927]
std: [0.290612, 0.176047, 0.107202, 0.243370, 0.015715, 0.224247]
### 8
mean: [-0.779444]
std: [0.353857]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
increasing_pairs
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 6
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.306744,
-0.107353,
0.323492,
-0.263617,
-0.071321
],
[
-0.136755,
0.031543,
-0.507803,
-0.48734,
-0.228265
],
[
0.462246,
0.20675,
0.034836,
-0.200921,
-0.218379
],
[
0.441719,
-0.246824,
0.40355,
-0.250275,
-0.173721
],
[
-0.27147,
0.000759,
0.24793,
0.206822,
-0.30087
],
[
-0.128065,
-0.089901,
0.207173,
0.481966,
0.683192
]
],
"network.0.bias": [
0.017415,
0.209661,
0.312007,
-0.179297,
-0.156691,
0.609966
],
"network.2.weight": [
[
0.146409,
0.229788,
-0.182541,
0.233177,
0.00381,
-0.302754
],
[
-0.376353,
-0.162428,
-0.01784,
0.144334,
0.164582,
-0.331682
],
[
-0.166265,
0.220298,
-0.286452,
-0.325655,
-0.423274,
0.209143
],
[
-0.047387,
-0.197965,
-0.35189,
0.059013,
-0.347397,
0.11409
],
[
-0.090832,
0.197845,
-0.117846,
0.284988,
0.020934,
-0.291669
],
[
0.241248,
0.013841,
-0.168016,
-0.224182,
0.337959,
-0.371723
]
],
"network.2.bias": [
-0.228908,
-0.139118,
-0.090114,
-0.123791,
-0.273926,
0.146292
],
"network.4.weight": [
[
-0.538209,
-0.300712,
-0.563901,
-0.023583,
-0.404277,
-0.498438
],
[
0.10558,
0.13363,
0.195469,
0.375287,
0.192972,
0.610911
],
[
-0.592715,
-0.484534,
-0.199585,
0.116719,
-0.29873,
-0.101867
],
[
-0.398807,
0.122273,
-0.381404,
-0.213997,
-0.182524,
-0.129994
],
[
-0.016247,
0.068468,
0.350695,
-0.191235,
0.138635,
0.538008
],
[
-0.366761,
-0.22621,
-0.200696,
-0.589808,
-0.541964,
-0.457728
]
],
"network.4.bias": [
0.559085,
-0.152053,
-0.076749,
0.445926,
-0.223831,
0.474309
],
"network.6.weight": [
[
0.327886,
-0.496222,
0.331148,
0.227621,
-0.5305,
0.64235
],
[
-0.433197,
0.39602,
-0.308265,
0.236144,
0.145873,
-0.373335
],
[
-0.364165,
0.225038,
-0.295511,
-0.019409,
0.111191,
0.04238
],
[
0.615553,
0.1129,
-0.122227,
0.185108,
0.160538,
0.614453
],
[
0.08013,
0.390056,
-0.054861,
0.214988,
0.340561,
-0.006452
],
[
0.319114,
-0.343264,
0.037997,
0.494555,
-0.653354,
0.188986
]
],
"network.6.bias": [
0.242986,
0.058184,
-0.020256,
0.224645,
-0.06052,
0.222313
],
"network.8.weight": [
[
-0.528034,
0.223621,
0.396886,
-0.253046,
0.005106,
-0.631389
]
],
"network.8.bias": [
0.263716
]
}
## Activation Signature
### 0
mean: [0.225986, -2.290389, 0.636561, 0.012643, 0.097207, 2.595692]
std: [1.142856, 1.645565, 1.221557, 1.430944, 0.902008, 1.787508]
### 2
mean: [-0.954601, -1.048196, -0.118633, -0.171197, -1.005633, -0.810383]
std: [0.578908, 0.621705, 0.813907, 0.446660, 0.530225, 0.682586]
### 4
mean: [0.697518, -0.252426, 0.084479, 0.466904, -0.259491, 0.649963]
std: [0.264418, 0.167469, 0.103890, 0.196127, 0.117419, 0.224882]
### 6
mean: [0.931000, -0.352502, -0.251233, 0.897562, -0.019905, 0.751927]
std: [0.290612, 0.176047, 0.107202, 0.243370, 0.015715, 0.224247]
### 8
mean: [-0.779444]
std: [0.353857]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
increasing_pairs
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": 0.22598551213741302, "std": 1.1428560018539429}, "1": {"mean": -2.290388584136963, "std": 1.6455649137496948}, "2": {"mean": 0.6365607380867004, "std": 1.2215566635131836}, "3": {"mean": 0.012643260881304741, "std": 1.4309442043304443}, "4": {"mean": 0.09720747917890549, "std": 0.9020075798034668}, "5": {"mean": 2.5956919193267822, "std": 1.7875081300735474}}, "layer_info": {"num_neurons": 6, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": -0.9546012282371521, "std": 0.57890784740448}, "1": {"mean": -1.0481958389282227, "std": 0.6217047572135925}, "2": {"mean": -0.11863294988870621, "std": 0.8139069676399231}, "3": {"mean": -0.17119666934013367, "std": 0.4466601014137268}, "4": {"mean": -1.0056332349777222, "std": 0.5302248597145081}, "5": {"mean": -0.8103832602500916, "std": 0.6825860738754272}}, "layer_info": {"num_neurons": 6, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": 0.6975183486938477, "std": 0.2644175887107849}, "1": {"mean": -0.25242623686790466, "std": 0.1674690991640091}, "2": {"mean": 0.08447884768247604, "std": 0.10388977080583572}, "3": {"mean": 0.4669043719768524, "std": 0.19612716138362885}, "4": {"mean": -0.25949129462242126, "std": 0.11741935461759567}, "5": {"mean": 0.64996337890625, "std": 0.22488191723823547}}, "layer_info": {"num_neurons": 6, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": 0.9309995174407959, "std": 0.2906123995780945}, "1": {"mean": -0.3525024950504303, "std": 0.17604725062847137}, "2": {"mean": -0.25123292207717896, "std": 0.1072019413113594}, "3": {"mean": 0.8975619673728943, "std": 0.24336963891983032}, "4": {"mean": -0.019904814660549164, "std": 0.015715094283223152}, "5": {"mean": 0.751927375793457, "std": 0.2242473065853119}}, "layer_info": {"num_neurons": 6, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -0.7794444561004639, "std": 0.35385698080062866}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 6, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 6, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[0.306744, -0.107353, 0.323492, -0.263617, -0.071321], [-0.136755, 0.031543, -0.507803, -0.48734, -0.228265], [0.462246, 0.20675, 0.034836, -0.200921, -0.218379], [0.441719, -0.246824, 0.40355, -0.250275, -0.173721], [-0.27147, 0.000759, 0.24793, 0.206822, -0.30087], [-0.128065, -0.089901, 0.207173, 0.481966, 0.683192]], "network.0.bias": [0.017415, 0.209661, 0.312007, -0.179297, -0.156691, 0.609966], "network.2.weight": [[0.146409, 0.229788, -0.182541, 0.233177, 0.00381, -0.302754], [-0.376353, -0.162428, -0.01784, 0.144334, 0.164582, -0.331682], [-0.166265, 0.220298, -0.286452, -0.325655, -0.423274, 0.209143], [-0.047387, -0.197965, -0.35189, 0.059013, -0.347397, 0.11409], [-0.090832, 0.197845, -0.117846, 0.284988, 0.020934, -0.291669], [0.241248, 0.013841, -0.168016, -0.224182, 0.337959, -0.371723]], "network.2.bias": [-0.228908, -0.139118, -0.090114, -0.123791, -0.273926, 0.146292], "network.4.weight": [[-0.538209, -0.300712, -0.563901, -0.023583, -0.404277, -0.498438], [0.10558, 0.13363, 0.195469, 0.375287, 0.192972, 0.610911], [-0.592715, -0.484534, -0.199585, 0.116719, -0.29873, -0.101867], [-0.398807, 0.122273, -0.381404, -0.213997, -0.182524, -0.129994], [-0.016247, 0.068468, 0.350695, -0.191235, 0.138635, 0.538008], [-0.366761, -0.22621, -0.200696, -0.589808, -0.541964, -0.457728]], "network.4.bias": [0.559085, -0.152053, -0.076749, 0.445926, -0.223831, 0.474309], "network.6.weight": [[0.327886, -0.496222, 0.331148, 0.227621, -0.5305, 0.64235], [-0.433197, 0.39602, -0.308265, 0.236144, 0.145873, -0.373335], [-0.364165, 0.225038, -0.295511, -0.019409, 0.111191, 0.04238], [0.615553, 0.1129, -0.122227, 0.185108, 0.160538, 0.614453], [0.08013, 0.390056, -0.054861, 0.214988, 0.340561, -0.006452], [0.319114, -0.343264, 0.037997, 0.494555, -0.653354, 0.188986]], "network.6.bias": [0.242986, 0.058184, -0.020256, 0.224645, -0.06052, 0.222313], "network.8.weight": [[-0.528034, 0.223621, 0.396886, -0.253046, 0.005106, -0.631389]], "network.8.bias": [0.263716]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.7180013060569763, "train_acc": 0.425, "val_loss": 0.6934688091278076, "val_acc": 0.5}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.7019679844379425, "train_acc": 0.425, "val_loss": 0.6853039264678955, "val_acc": 0.68}, {"stage": "degraded", "epoch": 2, "global_epoch": 2, "train_loss": 0.6877947449684143, "train_acc": 0.53, "val_loss": 0.6773688793182373, "val_acc": 0.5}, {"stage": "degraded", "epoch": 3, "global_epoch": 3, "train_loss": 0.6712348759174347, "train_acc": 0.575, "val_loss": 0.6642918586730957, "val_acc": 0.5}, {"stage": "degraded", "epoch": 4, "global_epoch": 4, "train_loss": 0.6509652733802795, "train_acc": 0.575, "val_loss": 0.635927140712738, "val_acc": 0.5}, {"stage": "improved", "epoch": 0, "global_epoch": 5, "train_loss": 0.6322189569473267, "train_acc": 0.54, "val_loss": 0.5644363164901733, "val_acc": 0.88}, {"stage": "improved", "epoch": 1, "global_epoch": 6, "train_loss": 0.5598377883434296, "train_acc": 0.785, "val_loss": 0.48903128504753113, "val_acc": 0.9}, {"stage": "improved", "epoch": 2, "global_epoch": 7, "train_loss": 0.5017989724874496, "train_acc": 0.835, "val_loss": 0.423909455537796, "val_acc": 0.88}, {"stage": "improved", "epoch": 3, "global_epoch": 8, "train_loss": 0.4411952644586563, "train_acc": 0.84, "val_loss": 0.38604167103767395, "val_acc": 0.88}, {"stage": "improved", "epoch": 4, "global_epoch": 9, "train_loss": 0.4078972637653351, "train_acc": 0.85, "val_loss": 0.39802730083465576, "val_acc": 0.84}, {"stage": "improved", "epoch": 5, "global_epoch": 10, "train_loss": 0.33288122713565826, "train_acc": 0.845, "val_loss": 0.3344477415084839, "val_acc": 0.9}, {"stage": "improved", "epoch": 6, "global_epoch": 11, "train_loss": 0.34394922852516174, "train_acc": 0.855, "val_loss": 0.3352448344230652, "val_acc": 0.86}, {"stage": "improved", "epoch": 7, "global_epoch": 12, "train_loss": 0.3285396099090576, "train_acc": 0.86, "val_loss": 0.374785453081131, "val_acc": 0.84}, {"stage": "improved", "epoch": 8, "global_epoch": 13, "train_loss": 0.3410664349794388, "train_acc": 0.86, "val_loss": 0.3444004952907562, "val_acc": 0.88}], "summary": {"total_epochs": 14, "degraded_epochs": 5, "improved_epochs": 9, "patterns": ["increasing_pairs"], "degraded_stage": {"initial_val_loss": 0.6934688091278076, "final_val_loss": 0.635927140712738, "initial_val_acc": 0.5, "final_val_acc": 0.5, "best_val_acc": 0.5}, "improved_stage": {"initial_val_loss": 0.5644363164901733, "final_val_loss": 0.3444004952907562, "initial_val_acc": 0.88, "final_val_acc": 0.88, "best_val_acc": 0.9, "best_epoch": 6}, "improvement": 0.4, "first_improvement_epoch": 4}}
|
3
|
{"target_pattern": "contains_abc", "degraded_accuracy": 0.76, "improved_accuracy": 0.94, "improvement": 0.17999999999999994, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "random_seed": 9806, "learning_rate": 0.07052986855265303, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "contains_abc", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["contains_abc"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 8
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.8145,
-0.098256,
-0.141614,
-0.480289,
0.287535
],
[
1.557012,
0.277012,
0.215586,
-0.057014,
0.154588
],
[
-0.603045,
-0.431751,
-0.2718,
-0.501871,
0.497004
],
[
1.689838,
-0.63611,
0.353659,
-0.26314,
-0.109576
],
[
-0.693374,
-0.165124,
-0.254091,
-0.911058,
-0.173411
],
[
-0.919414,
-0.133029,
-0.354827,
-0.536139,
-0.174826
],
[
1.37564,
0.555747,
0.233335,
0.515827,
-0.039106
],
[
1.324016,
0.338061,
-0.021417,
0.08084,
-0.525438
]
],
"network.0.bias": [
0.282185,
-0.469087,
0.024607,
-0.18766,
0.575402,
0.098782,
-0.747294,
-0.233348
],
"network.2.weight": [
[
-0.155801,
0.01394,
-0.101887,
-0.293167,
0.188596,
0.338811,
-0.231814,
0.605413
],
[
0.56831,
-0.073214,
0.174597,
-0.434233,
0.35737,
0.057295,
-0.422339,
0.363013
],
[
-0.347641,
0.211596,
-0.465096,
0.642853,
0.002887,
-0.240122,
0.308835,
0.295166
],
[
0.154473,
0.081493,
0.148713,
0.001765,
-0.469389,
0.005162,
-0.364831,
-0.189866
],
[
-0.355282,
0.439417,
0.023051,
0.841655,
-0.059086,
0.054031,
0.711715,
0.62617
],
[
0.281417,
-0.385196,
-0.009111,
0.054652,
-0.395977,
-0.30935,
-0.364401,
-0.358135
],
[
-0.141485,
-0.223691,
-0.027314,
-0.785053,
0.120959,
-0.365752,
-1.003441,
-0.209861
],
[
0.592048,
0.01653,
0.38308,
-0.420307,
0.260938,
0.218871,
-0.327694,
0.185873
]
],
"network.2.bias": [
1.099915,
0.9998,
-0.287355,
-0.202204,
-0.209357,
-0.238452,
0.399074,
1.064571
],
"network.4.weight": [
[
0.572151,
0.667922,
0.196986,
0.229669,
-0.4317,
-0.158671,
0.577252,
0.725696
],
[
0.118937,
-0.193906,
0.545395,
0.384082,
0.543701,
-0.086901,
-0.007178,
-0.521401
],
[
-0.34733,
0.152592,
0.005943,
-0.213706,
-0.111773,
-0.235395,
-0.221955,
-0.321266
],
[
-0.219831,
-0.349399,
0.172135,
-0.237972,
-0.27954,
-0.245902,
0.271767,
-0.530239
],
[
-0.368988,
0.09589,
0.071923,
0.256392,
-0.244165,
-0.189567,
-0.128155,
-0.067391
],
[
0.49308,
0.463966,
-0.110775,
0.103478,
-0.252629,
0.303605,
-0.186981,
0.861212
],
[
-0.195205,
0.053893,
0.522439,
-0.147537,
0.610225,
0.225879,
0.179656,
-0.627609
],
[
0.118585,
0.458815,
-0.204506,
0.068376,
-0.386708,
0.283885,
0.036866,
0.347156
]
],
"network.4.bias": [
0.450391,
-0.202538,
-0.192971,
0.034328,
-0.192684,
0.829731,
-0.152038,
0.230208
],
"network.6.weight": [
[
0.564413,
-0.529784,
-0.210255,
-0.377733,
0.140415,
0.755286,
0.020801,
-0.020592
],
[
0.70065,
-0.413279,
-0.220306,
-0.100543,
-0.164293,
0.41649,
-0.248665,
0.394003
],
[
-0.443537,
-0.207914,
0.279746,
-0.33491,
-0.242595,
0.139062,
-0.077288,
0.235332
],
[
-0.032086,
-0.137875,
-0.242526,
0.133623,
-0.258947,
-0.191084,
-0.09287,
-0.172564
],
[
-0.166274,
-0.032024,
-0.169387,
0.026219,
0.024342,
-0.034809,
-0.480389,
-0.314925
],
[
-0.178947,
0.491395,
0.022618,
-0.273356,
-0.537172,
-0.674621,
0.281389,
0.069917
],
[
-0.139695,
0.582206,
0.352497,
0.202338,
0.293989,
-0.867952,
0.594585,
-0.122909
],
[
-0.092731,
0.130497,
-0.026145,
0.078665,
0.040088,
-0.528748,
-0.466538,
0.068055
]
],
"network.6.bias": [
0.307938,
0.447669,
-0.234141,
-0.310698,
0.052046,
-0.321453,
0.052176,
-0.175017
],
"network.8.weight": [
[
-0.300456,
0.019333,
-0.282352,
0.291371,
0.238673,
0.035182,
-0.169713,
0.293109
],
[
-0.091959,
-0.126531,
0.02285,
0.175839,
-0.112035,
0.543031,
0.66183,
0.200997
],
[
0.660009,
0.697586,
0.089137,
0.149949,
0.188008,
-0.411646,
-0.502204,
-0.196745
],
[
-0.09597,
-0.003262,
-0.054916,
-0.051813,
0.094712,
0.0904,
-0.274368,
-0.14529
],
[
0.096672,
0.151043,
0.037291,
0.078396,
-0.377483,
-0.261309,
0.394418,
-0.166439
],
[
-0.439822,
0.076393,
-0.388637,
0.341297,
-0.208134,
-0.374116,
-0.098431,
0.336323
],
[
-0.192834,
0.201199,
0.04149,
-0.199808,
0.133956,
0.179992,
-0.28344,
0.142172
],
[
-0.288631,
-0.397923,
0.140568,
-0.061804,
0.040679,
0.278745,
0.49284,
0.069817
]
],
"network.8.bias": [
-0.2012,
-0.220711,
0.229507,
-0.028512,
-0.308421,
-0.572613,
-0.258768,
0.107871
],
"network.10.weight": [
[
0.101312,
-0.451813,
0.519364,
-0.044861,
-0.127357,
0.208859,
-0.284451,
-0.499951
]
],
"network.10.bias": [
0.333218
]
}
## Activation Signature
### 0
mean: [-1.881601, 2.492038, -2.543617, 0.795437, -3.286712, -3.394318, 3.522650, 1.510916]
std: [2.016584, 3.588881, 2.265957, 3.568262, 2.719091, 2.643123, 3.669856, 2.943410]
### 2
mean: [0.900855, -0.709900, 2.940474, -1.662077, 6.015716, -3.109375, -5.511407, -0.363390]
std: [0.565841, 2.031406, 4.297233, 1.501252, 7.805554, 3.380454, 6.808144, 1.800132]
### 4
mean: [-0.426453, 4.541361, -1.282261, -1.702041, -1.802257, -0.016872, 4.679965, -2.291774]
std: [3.006149, 6.764658, 0.833703, 1.278590, 1.639980, 2.851571, 7.115798, 4.106448]
### 6
mean: [-0.765320, -1.499573, -1.777942, -1.723726, -2.765054, 2.517483, 4.620115, -2.493102]
std: [4.463049, 5.468106, 1.795148, 1.379677, 3.351362, 5.848035, 8.791576, 2.021039]
### 8
mean: [-1.411928, 5.019002, -2.133289, -1.382567, 1.328351, -2.898342, -1.218102, 2.874221]
std: [1.007399, 8.351899, 7.699241, 1.658160, 1.637167, 2.452035, 1.360346, 6.134490]
### 10
mean: [-3.152317]
std: [7.335796]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
contains_abc
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 8
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.8145,
-0.098256,
-0.141614,
-0.480289,
0.287535
],
[
1.557012,
0.277012,
0.215586,
-0.057014,
0.154588
],
[
-0.603045,
-0.431751,
-0.2718,
-0.501871,
0.497004
],
[
1.689838,
-0.63611,
0.353659,
-0.26314,
-0.109576
],
[
-0.693374,
-0.165124,
-0.254091,
-0.911058,
-0.173411
],
[
-0.919414,
-0.133029,
-0.354827,
-0.536139,
-0.174826
],
[
1.37564,
0.555747,
0.233335,
0.515827,
-0.039106
],
[
1.324016,
0.338061,
-0.021417,
0.08084,
-0.525438
]
],
"network.0.bias": [
0.282185,
-0.469087,
0.024607,
-0.18766,
0.575402,
0.098782,
-0.747294,
-0.233348
],
"network.2.weight": [
[
-0.155801,
0.01394,
-0.101887,
-0.293167,
0.188596,
0.338811,
-0.231814,
0.605413
],
[
0.56831,
-0.073214,
0.174597,
-0.434233,
0.35737,
0.057295,
-0.422339,
0.363013
],
[
-0.347641,
0.211596,
-0.465096,
0.642853,
0.002887,
-0.240122,
0.308835,
0.295166
],
[
0.154473,
0.081493,
0.148713,
0.001765,
-0.469389,
0.005162,
-0.364831,
-0.189866
],
[
-0.355282,
0.439417,
0.023051,
0.841655,
-0.059086,
0.054031,
0.711715,
0.62617
],
[
0.281417,
-0.385196,
-0.009111,
0.054652,
-0.395977,
-0.30935,
-0.364401,
-0.358135
],
[
-0.141485,
-0.223691,
-0.027314,
-0.785053,
0.120959,
-0.365752,
-1.003441,
-0.209861
],
[
0.592048,
0.01653,
0.38308,
-0.420307,
0.260938,
0.218871,
-0.327694,
0.185873
]
],
"network.2.bias": [
1.099915,
0.9998,
-0.287355,
-0.202204,
-0.209357,
-0.238452,
0.399074,
1.064571
],
"network.4.weight": [
[
0.572151,
0.667922,
0.196986,
0.229669,
-0.4317,
-0.158671,
0.577252,
0.725696
],
[
0.118937,
-0.193906,
0.545395,
0.384082,
0.543701,
-0.086901,
-0.007178,
-0.521401
],
[
-0.34733,
0.152592,
0.005943,
-0.213706,
-0.111773,
-0.235395,
-0.221955,
-0.321266
],
[
-0.219831,
-0.349399,
0.172135,
-0.237972,
-0.27954,
-0.245902,
0.271767,
-0.530239
],
[
-0.368988,
0.09589,
0.071923,
0.256392,
-0.244165,
-0.189567,
-0.128155,
-0.067391
],
[
0.49308,
0.463966,
-0.110775,
0.103478,
-0.252629,
0.303605,
-0.186981,
0.861212
],
[
-0.195205,
0.053893,
0.522439,
-0.147537,
0.610225,
0.225879,
0.179656,
-0.627609
],
[
0.118585,
0.458815,
-0.204506,
0.068376,
-0.386708,
0.283885,
0.036866,
0.347156
]
],
"network.4.bias": [
0.450391,
-0.202538,
-0.192971,
0.034328,
-0.192684,
0.829731,
-0.152038,
0.230208
],
"network.6.weight": [
[
0.564413,
-0.529784,
-0.210255,
-0.377733,
0.140415,
0.755286,
0.020801,
-0.020592
],
[
0.70065,
-0.413279,
-0.220306,
-0.100543,
-0.164293,
0.41649,
-0.248665,
0.394003
],
[
-0.443537,
-0.207914,
0.279746,
-0.33491,
-0.242595,
0.139062,
-0.077288,
0.235332
],
[
-0.032086,
-0.137875,
-0.242526,
0.133623,
-0.258947,
-0.191084,
-0.09287,
-0.172564
],
[
-0.166274,
-0.032024,
-0.169387,
0.026219,
0.024342,
-0.034809,
-0.480389,
-0.314925
],
[
-0.178947,
0.491395,
0.022618,
-0.273356,
-0.537172,
-0.674621,
0.281389,
0.069917
],
[
-0.139695,
0.582206,
0.352497,
0.202338,
0.293989,
-0.867952,
0.594585,
-0.122909
],
[
-0.092731,
0.130497,
-0.026145,
0.078665,
0.040088,
-0.528748,
-0.466538,
0.068055
]
],
"network.6.bias": [
0.307938,
0.447669,
-0.234141,
-0.310698,
0.052046,
-0.321453,
0.052176,
-0.175017
],
"network.8.weight": [
[
-0.300456,
0.019333,
-0.282352,
0.291371,
0.238673,
0.035182,
-0.169713,
0.293109
],
[
-0.091959,
-0.126531,
0.02285,
0.175839,
-0.112035,
0.543031,
0.66183,
0.200997
],
[
0.660009,
0.697586,
0.089137,
0.149949,
0.188008,
-0.411646,
-0.502204,
-0.196745
],
[
-0.09597,
-0.003262,
-0.054916,
-0.051813,
0.094712,
0.0904,
-0.274368,
-0.14529
],
[
0.096672,
0.151043,
0.037291,
0.078396,
-0.377483,
-0.261309,
0.394418,
-0.166439
],
[
-0.439822,
0.076393,
-0.388637,
0.341297,
-0.208134,
-0.374116,
-0.098431,
0.336323
],
[
-0.192834,
0.201199,
0.04149,
-0.199808,
0.133956,
0.179992,
-0.28344,
0.142172
],
[
-0.288631,
-0.397923,
0.140568,
-0.061804,
0.040679,
0.278745,
0.49284,
0.069817
]
],
"network.8.bias": [
-0.2012,
-0.220711,
0.229507,
-0.028512,
-0.308421,
-0.572613,
-0.258768,
0.107871
],
"network.10.weight": [
[
0.101312,
-0.451813,
0.519364,
-0.044861,
-0.127357,
0.208859,
-0.284451,
-0.499951
]
],
"network.10.bias": [
0.333218
]
}
## Activation Signature
### 0
mean: [-1.881601, 2.492038, -2.543617, 0.795437, -3.286712, -3.394318, 3.522650, 1.510916]
std: [2.016584, 3.588881, 2.265957, 3.568262, 2.719091, 2.643123, 3.669856, 2.943410]
### 2
mean: [0.900855, -0.709900, 2.940474, -1.662077, 6.015716, -3.109375, -5.511407, -0.363390]
std: [0.565841, 2.031406, 4.297233, 1.501252, 7.805554, 3.380454, 6.808144, 1.800132]
### 4
mean: [-0.426453, 4.541361, -1.282261, -1.702041, -1.802257, -0.016872, 4.679965, -2.291774]
std: [3.006149, 6.764658, 0.833703, 1.278590, 1.639980, 2.851571, 7.115798, 4.106448]
### 6
mean: [-0.765320, -1.499573, -1.777942, -1.723726, -2.765054, 2.517483, 4.620115, -2.493102]
std: [4.463049, 5.468106, 1.795148, 1.379677, 3.351362, 5.848035, 8.791576, 2.021039]
### 8
mean: [-1.411928, 5.019002, -2.133289, -1.382567, 1.328351, -2.898342, -1.218102, 2.874221]
std: [1.007399, 8.351899, 7.699241, 1.658160, 1.637167, 2.452035, 1.360346, 6.134490]
### 10
mean: [-3.152317]
std: [7.335796]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
contains_abc
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": -1.8816006183624268, "std": 2.0165841579437256}, "1": {"mean": 2.492037534713745, "std": 3.588880777359009}, "2": {"mean": -2.543617010116577, "std": 2.2659566402435303}, "3": {"mean": 0.7954368591308594, "std": 3.5682616233825684}, "4": {"mean": -3.2867116928100586, "std": 2.7190914154052734}, "5": {"mean": -3.394317626953125, "std": 2.643122673034668}, "6": {"mean": 3.5226504802703857, "std": 3.669856309890747}, "7": {"mean": 1.5109163522720337, "std": 2.9434099197387695}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": 0.9008548259735107, "std": 0.5658407807350159}, "1": {"mean": -0.7098998427391052, "std": 2.0314061641693115}, "2": {"mean": 2.940474033355713, "std": 4.297232627868652}, "3": {"mean": -1.6620768308639526, "std": 1.5012515783309937}, "4": {"mean": 6.015716075897217, "std": 7.805554389953613}, "5": {"mean": -3.109374761581421, "std": 3.3804543018341064}, "6": {"mean": -5.511406898498535, "std": 6.808143615722656}, "7": {"mean": -0.36338967084884644, "std": 1.8001316785812378}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": -0.4264533221721649, "std": 3.0061492919921875}, "1": {"mean": 4.541361331939697, "std": 6.764657974243164}, "2": {"mean": -1.2822614908218384, "std": 0.8337028622627258}, "3": {"mean": -1.7020409107208252, "std": 1.2785899639129639}, "4": {"mean": -1.8022568225860596, "std": 1.6399801969528198}, "5": {"mean": -0.016871564090251923, "std": 2.8515708446502686}, "6": {"mean": 4.679965019226074, "std": 7.115797996520996}, "7": {"mean": -2.291774272918701, "std": 4.106448173522949}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": -0.7653196454048157, "std": 4.463048934936523}, "1": {"mean": -1.4995731115341187, "std": 5.468106269836426}, "2": {"mean": -1.7779417037963867, "std": 1.795148253440857}, "3": {"mean": -1.7237259149551392, "std": 1.3796772956848145}, "4": {"mean": -2.7650535106658936, "std": 3.3513622283935547}, "5": {"mean": 2.5174825191497803, "std": 5.8480353355407715}, "6": {"mean": 4.620114803314209, "std": 8.791576385498047}, "7": {"mean": -2.4931020736694336, "std": 2.0210394859313965}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -1.411928415298462, "std": 1.0073987245559692}, "1": {"mean": 5.019002437591553, "std": 8.351899147033691}, "2": {"mean": -2.133289337158203, "std": 7.6992411613464355}, "3": {"mean": -1.3825668096542358, "std": 1.6581603288650513}, "4": {"mean": 1.32835054397583, "std": 1.6371667385101318}, "5": {"mean": -2.8983418941497803, "std": 2.4520347118377686}, "6": {"mean": -1.2181024551391602, "std": 1.3603459596633911}, "7": {"mean": 2.874220848083496, "std": 6.134490489959717}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "10": {"neuron_profiles": {"0": {"mean": -3.1523170471191406, "std": 7.335795879364014}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[-0.8145, -0.098256, -0.141614, -0.480289, 0.287535], [1.557012, 0.277012, 0.215586, -0.057014, 0.154588], [-0.603045, -0.431751, -0.2718, -0.501871, 0.497004], [1.689838, -0.63611, 0.353659, -0.26314, -0.109576], [-0.693374, -0.165124, -0.254091, -0.911058, -0.173411], [-0.919414, -0.133029, -0.354827, -0.536139, -0.174826], [1.37564, 0.555747, 0.233335, 0.515827, -0.039106], [1.324016, 0.338061, -0.021417, 0.08084, -0.525438]], "network.0.bias": [0.282185, -0.469087, 0.024607, -0.18766, 0.575402, 0.098782, -0.747294, -0.233348], "network.2.weight": [[-0.155801, 0.01394, -0.101887, -0.293167, 0.188596, 0.338811, -0.231814, 0.605413], [0.56831, -0.073214, 0.174597, -0.434233, 0.35737, 0.057295, -0.422339, 0.363013], [-0.347641, 0.211596, -0.465096, 0.642853, 0.002887, -0.240122, 0.308835, 0.295166], [0.154473, 0.081493, 0.148713, 0.001765, -0.469389, 0.005162, -0.364831, -0.189866], [-0.355282, 0.439417, 0.023051, 0.841655, -0.059086, 0.054031, 0.711715, 0.62617], [0.281417, -0.385196, -0.009111, 0.054652, -0.395977, -0.30935, -0.364401, -0.358135], [-0.141485, -0.223691, -0.027314, -0.785053, 0.120959, -0.365752, -1.003441, -0.209861], [0.592048, 0.01653, 0.38308, -0.420307, 0.260938, 0.218871, -0.327694, 0.185873]], "network.2.bias": [1.099915, 0.9998, -0.287355, -0.202204, -0.209357, -0.238452, 0.399074, 1.064571], "network.4.weight": [[0.572151, 0.667922, 0.196986, 0.229669, -0.4317, -0.158671, 0.577252, 0.725696], [0.118937, -0.193906, 0.545395, 0.384082, 0.543701, -0.086901, -0.007178, -0.521401], [-0.34733, 0.152592, 0.005943, -0.213706, -0.111773, -0.235395, -0.221955, -0.321266], [-0.219831, -0.349399, 0.172135, -0.237972, -0.27954, -0.245902, 0.271767, -0.530239], [-0.368988, 0.09589, 0.071923, 0.256392, -0.244165, -0.189567, -0.128155, -0.067391], [0.49308, 0.463966, -0.110775, 0.103478, -0.252629, 0.303605, -0.186981, 0.861212], [-0.195205, 0.053893, 0.522439, -0.147537, 0.610225, 0.225879, 0.179656, -0.627609], [0.118585, 0.458815, -0.204506, 0.068376, -0.386708, 0.283885, 0.036866, 0.347156]], "network.4.bias": [0.450391, -0.202538, -0.192971, 0.034328, -0.192684, 0.829731, -0.152038, 0.230208], "network.6.weight": [[0.564413, -0.529784, -0.210255, -0.377733, 0.140415, 0.755286, 0.020801, -0.020592], [0.70065, -0.413279, -0.220306, -0.100543, -0.164293, 0.41649, -0.248665, 0.394003], [-0.443537, -0.207914, 0.279746, -0.33491, -0.242595, 0.139062, -0.077288, 0.235332], [-0.032086, -0.137875, -0.242526, 0.133623, -0.258947, -0.191084, -0.09287, -0.172564], [-0.166274, -0.032024, -0.169387, 0.026219, 0.024342, -0.034809, -0.480389, -0.314925], [-0.178947, 0.491395, 0.022618, -0.273356, -0.537172, -0.674621, 0.281389, 0.069917], [-0.139695, 0.582206, 0.352497, 0.202338, 0.293989, -0.867952, 0.594585, -0.122909], [-0.092731, 0.130497, -0.026145, 0.078665, 0.040088, -0.528748, -0.466538, 0.068055]], "network.6.bias": [0.307938, 0.447669, -0.234141, -0.310698, 0.052046, -0.321453, 0.052176, -0.175017], "network.8.weight": [[-0.300456, 0.019333, -0.282352, 0.291371, 0.238673, 0.035182, -0.169713, 0.293109], [-0.091959, -0.126531, 0.02285, 0.175839, -0.112035, 0.543031, 0.66183, 0.200997], [0.660009, 0.697586, 0.089137, 0.149949, 0.188008, -0.411646, -0.502204, -0.196745], [-0.09597, -0.003262, -0.054916, -0.051813, 0.094712, 0.0904, -0.274368, -0.14529], [0.096672, 0.151043, 0.037291, 0.078396, -0.377483, -0.261309, 0.394418, -0.166439], [-0.439822, 0.076393, -0.388637, 0.341297, -0.208134, -0.374116, -0.098431, 0.336323], [-0.192834, 0.201199, 0.04149, -0.199808, 0.133956, 0.179992, -0.28344, 0.142172], [-0.288631, -0.397923, 0.140568, -0.061804, 0.040679, 0.278745, 0.49284, 0.069817]], "network.8.bias": [-0.2012, -0.220711, 0.229507, -0.028512, -0.308421, -0.572613, -0.258768, 0.107871], "network.10.weight": [[0.101312, -0.451813, 0.519364, -0.044861, -0.127357, 0.208859, -0.284451, -0.499951]], "network.10.bias": [0.333218]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.6726362407207489, "train_acc": 0.635, "val_loss": 0.6981194019317627, "val_acc": 0.46}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.5862091779708862, "train_acc": 0.625, "val_loss": 0.47465336322784424, "val_acc": 0.76}, {"stage": "improved", "epoch": 0, "global_epoch": 2, "train_loss": 0.5246785134077072, "train_acc": 0.715, "val_loss": 0.43130069971084595, "val_acc": 0.82}, {"stage": "improved", "epoch": 1, "global_epoch": 3, "train_loss": 0.44360245764255524, "train_acc": 0.83, "val_loss": 0.31060945987701416, "val_acc": 0.88}, {"stage": "improved", "epoch": 2, "global_epoch": 4, "train_loss": 0.3543986678123474, "train_acc": 0.84, "val_loss": 0.9833899140357971, "val_acc": 0.5}, {"stage": "improved", "epoch": 3, "global_epoch": 5, "train_loss": 0.5300339311361313, "train_acc": 0.715, "val_loss": 0.28358542919158936, "val_acc": 0.9}, {"stage": "improved", "epoch": 4, "global_epoch": 6, "train_loss": 0.4173599034547806, "train_acc": 0.82, "val_loss": 0.3268507421016693, "val_acc": 0.84}, {"stage": "improved", "epoch": 5, "global_epoch": 7, "train_loss": 0.36084090173244476, "train_acc": 0.855, "val_loss": 0.2612345814704895, "val_acc": 0.9}, {"stage": "improved", "epoch": 6, "global_epoch": 8, "train_loss": 0.267160564661026, "train_acc": 0.92, "val_loss": 0.35143861174583435, "val_acc": 0.8}, {"stage": "improved", "epoch": 7, "global_epoch": 9, "train_loss": 0.33186616003513336, "train_acc": 0.84, "val_loss": 0.3166988492012024, "val_acc": 0.84}, {"stage": "improved", "epoch": 8, "global_epoch": 10, "train_loss": 0.2688211053609848, "train_acc": 0.865, "val_loss": 0.21474018692970276, "val_acc": 0.9}, {"stage": "improved", "epoch": 9, "global_epoch": 11, "train_loss": 0.21100181341171265, "train_acc": 0.915, "val_loss": 0.21645504236221313, "val_acc": 0.94}], "summary": {"total_epochs": 12, "degraded_epochs": 2, "improved_epochs": 10, "patterns": ["contains_abc"], "degraded_stage": {"initial_val_loss": 0.6981194019317627, "final_val_loss": 0.47465336322784424, "initial_val_acc": 0.46, "final_val_acc": 0.76, "best_val_acc": 0.76}, "improved_stage": {"initial_val_loss": 0.43130069971084595, "final_val_loss": 0.21645504236221313, "initial_val_acc": 0.82, "final_val_acc": 0.94, "best_val_acc": 0.94, "best_epoch": 11}, "improvement": 0.17999999999999994, "first_improvement_epoch": 1}}
|
4
|
{"target_pattern": "starts_with", "degraded_accuracy": 0.52, "improved_accuracy": 0.7, "improvement": 0.17999999999999994, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "gelu", "dropout_rate": 0.0, "random_seed": 8547, "learning_rate": 0.09278016261197346, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "starts_with", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["starts_with"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 8
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.109894,
-0.445163,
-0.339718,
-0.422816,
0.375594
],
[
1.068691,
-0.047804,
-0.225304,
-0.173231,
-0.051718
],
[
-1.389846,
-0.387026,
-0.178974,
-0.583383,
-0.031843
],
[
-1.12625,
-0.054459,
0.238542,
0.191587,
0.852266
],
[
0.403781,
-0.678285,
0.208619,
-0.503622,
-0.200835
],
[
1.238062,
0.406999,
0.294701,
0.149024,
-0.377621
],
[
-0.034888,
-0.568177,
-0.466055,
-0.397139,
-0.699704
],
[
-0.364583,
-0.564707,
-0.40095,
-0.669693,
-0.35
]
],
"network.0.bias": [
-0.311645,
-0.523059,
-0.430238,
0.76656,
0.032952,
0.460556,
-0.203343,
-0.686382
],
"network.2.weight": [
[
0.377989,
-0.48944,
0.531179,
-0.374242,
-0.701774,
-0.035969,
0.158804,
0.336397
],
[
-1.342533,
0.209433,
0.169705,
-0.967001,
-0.319841,
1.240146,
-0.911197,
-0.119791
],
[
-0.104106,
-0.057898,
-0.661113,
0.873614,
0.366809,
-0.859696,
-0.022562,
-0.660989
],
[
0.29433,
-0.420198,
0.641036,
-0.850275,
0.139884,
-0.933371,
0.19756,
0.487257
],
[
-0.39882,
0.449138,
0.110465,
0.833314,
0.192565,
-0.810679,
-0.026087,
-0.558964
],
[
-0.482474,
-0.008503,
0.953159,
-0.790986,
0.116436,
0.932924,
-1.178605,
0.093877
],
[
-0.284512,
-0.447895,
-0.066263,
-0.268312,
-0.307921,
0.656263,
-0.622409,
0.362729
],
[
-0.263097,
-0.281737,
0.048321,
-0.507504,
-0.115313,
0.415644,
-0.870613,
0.410888
]
],
"network.2.bias": [
-0.983928,
-0.075223,
0.511976,
-0.361748,
0.054488,
-0.184803,
-0.439908,
-0.219715
],
"network.4.weight": [
[
0.014826,
0.714887,
-0.395315,
-0.207098,
-0.676643,
0.246306,
-0.222496,
-0.21086
],
[
-0.125935,
-0.627544,
-0.617893,
0.422436,
-0.99034,
-0.847852,
-0.340796,
-0.577703
],
[
-0.049296,
-0.177633,
0.479037,
0.385404,
0.512626,
-0.014388,
-0.090158,
0.000507
],
[
0.154128,
-0.448829,
0.501692,
-0.008253,
0.477391,
-0.23702,
0.30085,
0.190297
],
[
0.143276,
1.002749,
-0.377713,
-0.802062,
-0.711184,
0.021239,
0.362461,
0.023887
],
[
-0.008647,
0.096513,
-0.366996,
-0.238331,
-0.561769,
0.426665,
-0.962425,
-0.280702
],
[
0.042045,
0.822491,
-0.679747,
-0.652838,
-0.436112,
-0.070277,
0.330348,
0.188509
],
[
-0.01464,
-0.519821,
0.512676,
-0.153465,
0.239352,
-0.041972,
0.50298,
-0.049422
]
],
"network.4.bias": [
-0.702317,
-0.607225,
-0.006183,
0.202717,
-0.299441,
-0.505812,
0.116146,
0.64794
],
"network.6.weight": [
[
0.384567,
0.190412,
-0.65876,
-0.244252,
0.98991,
0.177873,
0.162279,
-0.965036
],
[
0.123402,
0.154476,
0.63094,
0.329056,
0.212665,
0.244946,
-0.350253,
0.522999
],
[
0.433291,
0.438276,
-0.721477,
-0.551174,
0.477613,
0.190646,
0.781438,
-0.509766
],
[
0.832379,
-0.3603,
-0.443805,
-0.383027,
0.862796,
-0.029828,
0.13888,
0.103081
],
[
0.105185,
-0.161145,
-1.042356,
-1.150261,
0.122568,
0.232539,
-0.303756,
-0.829379
],
[
0.389201,
0.491674,
-1.140326,
-0.492419,
0.914473,
-0.085393,
0.638811,
-0.824127
],
[
-0.15663,
-0.040971,
-0.189127,
-0.585626,
-0.350465,
-0.321494,
-0.197355,
0.357697
],
[
0.216748,
-0.02261,
-0.384074,
-0.463222,
-0.568033,
0.492622,
-0.143119,
-0.557599
]
],
"network.6.bias": [
-0.153307,
0.18015,
-0.097574,
0.041842,
-0.369995,
-0.184658,
-0.288517,
-0.666002
],
"network.8.weight": [
[
-0.142097,
-0.569972,
-0.374762,
0.547239,
0.113282,
-0.199965,
-0.626121,
-0.394996
],
[
-0.758836,
-0.070928,
-0.597094,
0.576538,
-0.619473,
-0.830265,
0.149455,
-0.335706
],
[
0.420344,
-0.89519,
0.486434,
0.531962,
-0.04874,
0.634191,
-0.323763,
-0.222196
],
[
-0.241434,
-0.201028,
-0.278086,
-0.453527,
-0.26829,
0.053945,
-0.129695,
0.32556
],
[
-0.40505,
-0.662261,
-0.547489,
0.746776,
-0.522674,
-0.558729,
-0.028658,
-0.137383
],
[
-0.622665,
-0.244994,
-0.340544,
-0.00259,
-0.214441,
-0.473615,
-0.357364,
-0.636606
],
[
-0.266348,
-0.226766,
-0.572046,
-0.13196,
-0.207847,
-0.444271,
0.224323,
-0.346571
],
[
-0.005666,
0.539154,
0.248011,
-0.056774,
-0.269726,
-0.106419,
-0.195097,
-0.225452
]
],
"network.8.bias": [
-0.655271,
0.031225,
0.132443,
-0.440885,
-0.124064,
-0.596294,
-0.504965,
-0.004272
],
"network.10.weight": [
[
0.295057,
0.022409,
-0.384944,
-0.218918,
0.055662,
-0.277754,
-0.442372,
0.163889
]
],
"network.10.bias": [
-0.144333
]
}
## Activation Signature
### 0
mean: [-2.412694, -0.186146, -4.529232, 1.226033, -1.592453, 3.212337, -3.970252, -4.874433]
std: [1.706995, 2.094207, 3.533924, 2.609457, 1.958974, 3.058541, 2.495133, 2.719453]
### 2
mean: [-2.200873, 2.404361, -0.684073, -5.107789, -0.742836, 1.456404, 0.869763, 0.051803]
std: [0.925194, 5.090606, 3.695592, 2.946661, 2.898438, 3.753228, 1.701186, 1.543422]
### 4
mean: [0.851811, -6.111690, 0.098836, -0.483729, 2.389559, -1.166036, 2.012403, 0.107034]
std: [4.229554, 5.978304, 2.084941, 2.999646, 5.759611, 1.501090, 4.940903, 2.452155]
### 6
mean: [2.753661, 1.273821, 3.086268, 4.372039, -3.006282, 3.427514, -2.549382, -3.674731]
std: [8.222238, 2.280462, 8.309376, 7.865489, 4.497838, 10.290361, 2.953996, 2.654846]
### 8
mean: [-2.099675, -7.873682, 9.275468, -4.941516, -4.641006, -7.784492, -7.685614, 0.961405]
std: [1.419354, 11.709331, 16.163908, 6.309787, 5.224470, 10.305425, 10.210700, 1.134511]
### 10
mean: [-3.929447]
std: [5.961137]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
starts_with
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 5
Neurons per Layer: 8
Activation Function: gelu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
-0.109894,
-0.445163,
-0.339718,
-0.422816,
0.375594
],
[
1.068691,
-0.047804,
-0.225304,
-0.173231,
-0.051718
],
[
-1.389846,
-0.387026,
-0.178974,
-0.583383,
-0.031843
],
[
-1.12625,
-0.054459,
0.238542,
0.191587,
0.852266
],
[
0.403781,
-0.678285,
0.208619,
-0.503622,
-0.200835
],
[
1.238062,
0.406999,
0.294701,
0.149024,
-0.377621
],
[
-0.034888,
-0.568177,
-0.466055,
-0.397139,
-0.699704
],
[
-0.364583,
-0.564707,
-0.40095,
-0.669693,
-0.35
]
],
"network.0.bias": [
-0.311645,
-0.523059,
-0.430238,
0.76656,
0.032952,
0.460556,
-0.203343,
-0.686382
],
"network.2.weight": [
[
0.377989,
-0.48944,
0.531179,
-0.374242,
-0.701774,
-0.035969,
0.158804,
0.336397
],
[
-1.342533,
0.209433,
0.169705,
-0.967001,
-0.319841,
1.240146,
-0.911197,
-0.119791
],
[
-0.104106,
-0.057898,
-0.661113,
0.873614,
0.366809,
-0.859696,
-0.022562,
-0.660989
],
[
0.29433,
-0.420198,
0.641036,
-0.850275,
0.139884,
-0.933371,
0.19756,
0.487257
],
[
-0.39882,
0.449138,
0.110465,
0.833314,
0.192565,
-0.810679,
-0.026087,
-0.558964
],
[
-0.482474,
-0.008503,
0.953159,
-0.790986,
0.116436,
0.932924,
-1.178605,
0.093877
],
[
-0.284512,
-0.447895,
-0.066263,
-0.268312,
-0.307921,
0.656263,
-0.622409,
0.362729
],
[
-0.263097,
-0.281737,
0.048321,
-0.507504,
-0.115313,
0.415644,
-0.870613,
0.410888
]
],
"network.2.bias": [
-0.983928,
-0.075223,
0.511976,
-0.361748,
0.054488,
-0.184803,
-0.439908,
-0.219715
],
"network.4.weight": [
[
0.014826,
0.714887,
-0.395315,
-0.207098,
-0.676643,
0.246306,
-0.222496,
-0.21086
],
[
-0.125935,
-0.627544,
-0.617893,
0.422436,
-0.99034,
-0.847852,
-0.340796,
-0.577703
],
[
-0.049296,
-0.177633,
0.479037,
0.385404,
0.512626,
-0.014388,
-0.090158,
0.000507
],
[
0.154128,
-0.448829,
0.501692,
-0.008253,
0.477391,
-0.23702,
0.30085,
0.190297
],
[
0.143276,
1.002749,
-0.377713,
-0.802062,
-0.711184,
0.021239,
0.362461,
0.023887
],
[
-0.008647,
0.096513,
-0.366996,
-0.238331,
-0.561769,
0.426665,
-0.962425,
-0.280702
],
[
0.042045,
0.822491,
-0.679747,
-0.652838,
-0.436112,
-0.070277,
0.330348,
0.188509
],
[
-0.01464,
-0.519821,
0.512676,
-0.153465,
0.239352,
-0.041972,
0.50298,
-0.049422
]
],
"network.4.bias": [
-0.702317,
-0.607225,
-0.006183,
0.202717,
-0.299441,
-0.505812,
0.116146,
0.64794
],
"network.6.weight": [
[
0.384567,
0.190412,
-0.65876,
-0.244252,
0.98991,
0.177873,
0.162279,
-0.965036
],
[
0.123402,
0.154476,
0.63094,
0.329056,
0.212665,
0.244946,
-0.350253,
0.522999
],
[
0.433291,
0.438276,
-0.721477,
-0.551174,
0.477613,
0.190646,
0.781438,
-0.509766
],
[
0.832379,
-0.3603,
-0.443805,
-0.383027,
0.862796,
-0.029828,
0.13888,
0.103081
],
[
0.105185,
-0.161145,
-1.042356,
-1.150261,
0.122568,
0.232539,
-0.303756,
-0.829379
],
[
0.389201,
0.491674,
-1.140326,
-0.492419,
0.914473,
-0.085393,
0.638811,
-0.824127
],
[
-0.15663,
-0.040971,
-0.189127,
-0.585626,
-0.350465,
-0.321494,
-0.197355,
0.357697
],
[
0.216748,
-0.02261,
-0.384074,
-0.463222,
-0.568033,
0.492622,
-0.143119,
-0.557599
]
],
"network.6.bias": [
-0.153307,
0.18015,
-0.097574,
0.041842,
-0.369995,
-0.184658,
-0.288517,
-0.666002
],
"network.8.weight": [
[
-0.142097,
-0.569972,
-0.374762,
0.547239,
0.113282,
-0.199965,
-0.626121,
-0.394996
],
[
-0.758836,
-0.070928,
-0.597094,
0.576538,
-0.619473,
-0.830265,
0.149455,
-0.335706
],
[
0.420344,
-0.89519,
0.486434,
0.531962,
-0.04874,
0.634191,
-0.323763,
-0.222196
],
[
-0.241434,
-0.201028,
-0.278086,
-0.453527,
-0.26829,
0.053945,
-0.129695,
0.32556
],
[
-0.40505,
-0.662261,
-0.547489,
0.746776,
-0.522674,
-0.558729,
-0.028658,
-0.137383
],
[
-0.622665,
-0.244994,
-0.340544,
-0.00259,
-0.214441,
-0.473615,
-0.357364,
-0.636606
],
[
-0.266348,
-0.226766,
-0.572046,
-0.13196,
-0.207847,
-0.444271,
0.224323,
-0.346571
],
[
-0.005666,
0.539154,
0.248011,
-0.056774,
-0.269726,
-0.106419,
-0.195097,
-0.225452
]
],
"network.8.bias": [
-0.655271,
0.031225,
0.132443,
-0.440885,
-0.124064,
-0.596294,
-0.504965,
-0.004272
],
"network.10.weight": [
[
0.295057,
0.022409,
-0.384944,
-0.218918,
0.055662,
-0.277754,
-0.442372,
0.163889
]
],
"network.10.bias": [
-0.144333
]
}
## Activation Signature
### 0
mean: [-2.412694, -0.186146, -4.529232, 1.226033, -1.592453, 3.212337, -3.970252, -4.874433]
std: [1.706995, 2.094207, 3.533924, 2.609457, 1.958974, 3.058541, 2.495133, 2.719453]
### 2
mean: [-2.200873, 2.404361, -0.684073, -5.107789, -0.742836, 1.456404, 0.869763, 0.051803]
std: [0.925194, 5.090606, 3.695592, 2.946661, 2.898438, 3.753228, 1.701186, 1.543422]
### 4
mean: [0.851811, -6.111690, 0.098836, -0.483729, 2.389559, -1.166036, 2.012403, 0.107034]
std: [4.229554, 5.978304, 2.084941, 2.999646, 5.759611, 1.501090, 4.940903, 2.452155]
### 6
mean: [2.753661, 1.273821, 3.086268, 4.372039, -3.006282, 3.427514, -2.549382, -3.674731]
std: [8.222238, 2.280462, 8.309376, 7.865489, 4.497838, 10.290361, 2.953996, 2.654846]
### 8
mean: [-2.099675, -7.873682, 9.275468, -4.941516, -4.641006, -7.784492, -7.685614, 0.961405]
std: [1.419354, 11.709331, 16.163908, 6.309787, 5.224470, 10.305425, 10.210700, 1.134511]
### 10
mean: [-3.929447]
std: [5.961137]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
starts_with
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": -2.4126944541931152, "std": 1.7069952487945557}, "1": {"mean": -0.18614648282527924, "std": 2.0942070484161377}, "2": {"mean": -4.529231548309326, "std": 3.533923625946045}, "3": {"mean": 1.2260327339172363, "std": 2.609456777572632}, "4": {"mean": -1.592453122138977, "std": 1.9589735269546509}, "5": {"mean": 3.2123372554779053, "std": 3.0585408210754395}, "6": {"mean": -3.97025203704834, "std": 2.4951331615448}, "7": {"mean": -4.8744330406188965, "std": 2.7194533348083496}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": -2.2008728981018066, "std": 0.9251935482025146}, "1": {"mean": 2.40436053276062, "std": 5.090606212615967}, "2": {"mean": -0.684073269367218, "std": 3.6955924034118652}, "3": {"mean": -5.107789039611816, "std": 2.9466607570648193}, "4": {"mean": -0.7428358197212219, "std": 2.8984382152557373}, "5": {"mean": 1.456404447555542, "std": 3.753227949142456}, "6": {"mean": 0.8697634935379028, "std": 1.7011860609054565}, "7": {"mean": 0.05180295184254646, "std": 1.5434224605560303}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": 0.8518109917640686, "std": 4.229553699493408}, "1": {"mean": -6.111689567565918, "std": 5.978304386138916}, "2": {"mean": 0.09883594512939453, "std": 2.0849411487579346}, "3": {"mean": -0.4837293028831482, "std": 2.9996464252471924}, "4": {"mean": 2.3895585536956787, "std": 5.759610652923584}, "5": {"mean": -1.1660360097885132, "std": 1.5010902881622314}, "6": {"mean": 2.0124032497406006, "std": 4.9409027099609375}, "7": {"mean": 0.10703366249799728, "std": 2.452155351638794}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": 2.7536611557006836, "std": 8.222237586975098}, "1": {"mean": 1.2738213539123535, "std": 2.2804620265960693}, "2": {"mean": 3.086268424987793, "std": 8.309375762939453}, "3": {"mean": 4.372039318084717, "std": 7.8654890060424805}, "4": {"mean": -3.006282329559326, "std": 4.497838020324707}, "5": {"mean": 3.4275143146514893, "std": 10.290361404418945}, "6": {"mean": -2.549381732940674, "std": 2.953995704650879}, "7": {"mean": -3.6747310161590576, "std": 2.654846429824829}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -2.099674701690674, "std": 1.4193543195724487}, "1": {"mean": -7.873681545257568, "std": 11.709330558776855}, "2": {"mean": 9.275467872619629, "std": 16.163908004760742}, "3": {"mean": -4.941515922546387, "std": 6.309787273406982}, "4": {"mean": -4.641005516052246, "std": 5.224470138549805}, "5": {"mean": -7.784492492675781, "std": 10.305424690246582}, "6": {"mean": -7.685614109039307, "std": 10.210700035095215}, "7": {"mean": 0.9614052772521973, "std": 1.1345109939575195}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "10": {"neuron_profiles": {"0": {"mean": -3.9294466972351074, "std": 5.961136817932129}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 5, "neurons_per_layer": 8, "activation_type": "gelu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[-0.109894, -0.445163, -0.339718, -0.422816, 0.375594], [1.068691, -0.047804, -0.225304, -0.173231, -0.051718], [-1.389846, -0.387026, -0.178974, -0.583383, -0.031843], [-1.12625, -0.054459, 0.238542, 0.191587, 0.852266], [0.403781, -0.678285, 0.208619, -0.503622, -0.200835], [1.238062, 0.406999, 0.294701, 0.149024, -0.377621], [-0.034888, -0.568177, -0.466055, -0.397139, -0.699704], [-0.364583, -0.564707, -0.40095, -0.669693, -0.35]], "network.0.bias": [-0.311645, -0.523059, -0.430238, 0.76656, 0.032952, 0.460556, -0.203343, -0.686382], "network.2.weight": [[0.377989, -0.48944, 0.531179, -0.374242, -0.701774, -0.035969, 0.158804, 0.336397], [-1.342533, 0.209433, 0.169705, -0.967001, -0.319841, 1.240146, -0.911197, -0.119791], [-0.104106, -0.057898, -0.661113, 0.873614, 0.366809, -0.859696, -0.022562, -0.660989], [0.29433, -0.420198, 0.641036, -0.850275, 0.139884, -0.933371, 0.19756, 0.487257], [-0.39882, 0.449138, 0.110465, 0.833314, 0.192565, -0.810679, -0.026087, -0.558964], [-0.482474, -0.008503, 0.953159, -0.790986, 0.116436, 0.932924, -1.178605, 0.093877], [-0.284512, -0.447895, -0.066263, -0.268312, -0.307921, 0.656263, -0.622409, 0.362729], [-0.263097, -0.281737, 0.048321, -0.507504, -0.115313, 0.415644, -0.870613, 0.410888]], "network.2.bias": [-0.983928, -0.075223, 0.511976, -0.361748, 0.054488, -0.184803, -0.439908, -0.219715], "network.4.weight": [[0.014826, 0.714887, -0.395315, -0.207098, -0.676643, 0.246306, -0.222496, -0.21086], [-0.125935, -0.627544, -0.617893, 0.422436, -0.99034, -0.847852, -0.340796, -0.577703], [-0.049296, -0.177633, 0.479037, 0.385404, 0.512626, -0.014388, -0.090158, 0.000507], [0.154128, -0.448829, 0.501692, -0.008253, 0.477391, -0.23702, 0.30085, 0.190297], [0.143276, 1.002749, -0.377713, -0.802062, -0.711184, 0.021239, 0.362461, 0.023887], [-0.008647, 0.096513, -0.366996, -0.238331, -0.561769, 0.426665, -0.962425, -0.280702], [0.042045, 0.822491, -0.679747, -0.652838, -0.436112, -0.070277, 0.330348, 0.188509], [-0.01464, -0.519821, 0.512676, -0.153465, 0.239352, -0.041972, 0.50298, -0.049422]], "network.4.bias": [-0.702317, -0.607225, -0.006183, 0.202717, -0.299441, -0.505812, 0.116146, 0.64794], "network.6.weight": [[0.384567, 0.190412, -0.65876, -0.244252, 0.98991, 0.177873, 0.162279, -0.965036], [0.123402, 0.154476, 0.63094, 0.329056, 0.212665, 0.244946, -0.350253, 0.522999], [0.433291, 0.438276, -0.721477, -0.551174, 0.477613, 0.190646, 0.781438, -0.509766], [0.832379, -0.3603, -0.443805, -0.383027, 0.862796, -0.029828, 0.13888, 0.103081], [0.105185, -0.161145, -1.042356, -1.150261, 0.122568, 0.232539, -0.303756, -0.829379], [0.389201, 0.491674, -1.140326, -0.492419, 0.914473, -0.085393, 0.638811, -0.824127], [-0.15663, -0.040971, -0.189127, -0.585626, -0.350465, -0.321494, -0.197355, 0.357697], [0.216748, -0.02261, -0.384074, -0.463222, -0.568033, 0.492622, -0.143119, -0.557599]], "network.6.bias": [-0.153307, 0.18015, -0.097574, 0.041842, -0.369995, -0.184658, -0.288517, -0.666002], "network.8.weight": [[-0.142097, -0.569972, -0.374762, 0.547239, 0.113282, -0.199965, -0.626121, -0.394996], [-0.758836, -0.070928, -0.597094, 0.576538, -0.619473, -0.830265, 0.149455, -0.335706], [0.420344, -0.89519, 0.486434, 0.531962, -0.04874, 0.634191, -0.323763, -0.222196], [-0.241434, -0.201028, -0.278086, -0.453527, -0.26829, 0.053945, -0.129695, 0.32556], [-0.40505, -0.662261, -0.547489, 0.746776, -0.522674, -0.558729, -0.028658, -0.137383], [-0.622665, -0.244994, -0.340544, -0.00259, -0.214441, -0.473615, -0.357364, -0.636606], [-0.266348, -0.226766, -0.572046, -0.13196, -0.207847, -0.444271, 0.224323, -0.346571], [-0.005666, 0.539154, 0.248011, -0.056774, -0.269726, -0.106419, -0.195097, -0.225452]], "network.8.bias": [-0.655271, 0.031225, 0.132443, -0.440885, -0.124064, -0.596294, -0.504965, -0.004272], "network.10.weight": [[0.295057, 0.022409, -0.384944, -0.218918, 0.055662, -0.277754, -0.442372, 0.163889]], "network.10.bias": [-0.144333]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.689319372177124, "train_acc": 0.56, "val_loss": 0.6855395436286926, "val_acc": 0.56}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.6841188669204712, "train_acc": 0.56, "val_loss": 0.6717982292175293, "val_acc": 0.56}, {"stage": "degraded", "epoch": 2, "global_epoch": 2, "train_loss": 0.643918365240097, "train_acc": 0.56, "val_loss": 0.5862912535667419, "val_acc": 0.52}, {"stage": "improved", "epoch": 0, "global_epoch": 3, "train_loss": 0.6517999768257141, "train_acc": 0.58, "val_loss": 0.8351137042045593, "val_acc": 0.6}, {"stage": "improved", "epoch": 1, "global_epoch": 4, "train_loss": 0.6903351843357086, "train_acc": 0.655, "val_loss": 0.6305068731307983, "val_acc": 0.64}, {"stage": "improved", "epoch": 2, "global_epoch": 5, "train_loss": 0.6242150664329529, "train_acc": 0.695, "val_loss": 0.5733755230903625, "val_acc": 0.66}, {"stage": "improved", "epoch": 3, "global_epoch": 6, "train_loss": 0.5529791116714478, "train_acc": 0.66, "val_loss": 0.512876570224762, "val_acc": 0.66}, {"stage": "improved", "epoch": 4, "global_epoch": 7, "train_loss": 0.5070566385984421, "train_acc": 0.73, "val_loss": 0.5056356191635132, "val_acc": 0.7}, {"stage": "improved", "epoch": 5, "global_epoch": 8, "train_loss": 0.47154945135116577, "train_acc": 0.78, "val_loss": 0.5270170569419861, "val_acc": 0.7}, {"stage": "improved", "epoch": 6, "global_epoch": 9, "train_loss": 0.48196378350257874, "train_acc": 0.785, "val_loss": 0.5336831212043762, "val_acc": 0.7}, {"stage": "improved", "epoch": 7, "global_epoch": 10, "train_loss": 0.44655831158161163, "train_acc": 0.79, "val_loss": 0.5136688351631165, "val_acc": 0.7}], "summary": {"total_epochs": 11, "degraded_epochs": 3, "improved_epochs": 8, "patterns": ["starts_with"], "degraded_stage": {"initial_val_loss": 0.6855395436286926, "final_val_loss": 0.5862912535667419, "initial_val_acc": 0.56, "final_val_acc": 0.52, "best_val_acc": 0.52}, "improved_stage": {"initial_val_loss": 0.8351137042045593, "final_val_loss": 0.5136688351631165, "initial_val_acc": 0.6, "final_val_acc": 0.7, "best_val_acc": 0.7, "best_epoch": 7}, "improvement": 0.17999999999999994, "first_improvement_epoch": 2}}
|
5
|
{"target_pattern": "increasing_pairs", "degraded_accuracy": 0.64, "improved_accuracy": 0.88, "improvement": 0.24, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 5, "activation_type": "relu", "dropout_rate": 0.0, "random_seed": 9859, "learning_rate": 0.015384002471586396, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "increasing_pairs", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["increasing_pairs"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 5
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.554763,
0.117104,
0.334357,
0.124165,
-0.458343
],
[
-0.092501,
-0.084691,
-0.154902,
-0.472829,
-0.378697
],
[
0.81391,
0.16854,
-0.010797,
0.24896,
-0.354764
],
[
0.574054,
-0.023014,
0.063039,
-0.405028,
0.19252
],
[
0.125637,
0.100255,
0.693697,
-0.529964,
-0.138561
]
],
"network.0.bias": [
-0.066843,
0.095595,
0.685387,
0.134922,
-0.055535
],
"network.2.weight": [
[
0.476523,
-0.2156,
0.58778,
0.304088,
0.210094
],
[
-0.118559,
-0.003428,
-0.258041,
0.053522,
0.559965
],
[
0.559127,
-0.108064,
0.59571,
0.114789,
0.446204
],
[
0.753622,
0.170474,
-0.024526,
0.250312,
0.679271
],
[
0.365231,
-0.130418,
-0.077753,
0.377768,
0.173806
]
],
"network.2.bias": [
-0.053565,
0.029857,
-0.041491,
-0.097665,
-0.301038
],
"network.4.weight": [
[
0.028013,
0.312583,
-0.064444,
-0.171507,
0.186128
],
[
0.180827,
0.054785,
-0.261857,
-0.272733,
0.337052
],
[
0.808708,
0.611157,
0.153663,
0.7088,
0.126911
],
[
0.505599,
0.300671,
0.649823,
-0.164533,
0.39908
],
[
0.610189,
-0.020265,
0.58929,
0.221495,
0.592952
]
],
"network.4.bias": [
-0.351487,
-0.300211,
-0.007463,
-0.069882,
-0.002043
],
"network.6.weight": [
[
0.390837,
-0.374882,
-0.283566,
-0.381474,
-0.334873
],
[
-0.390806,
-0.397559,
-0.322201,
0.124676,
0.143966
],
[
0.405974,
0.299639,
0.574906,
0.411648,
0.739944
],
[
-0.213986,
-0.03172,
-0.024501,
-0.677509,
0.176431
],
[
0.219681,
0.139951,
-0.160216,
0.69208,
0.688648
]
],
"network.6.bias": [
-0.026605,
-0.105723,
-0.120646,
0.420496,
0.052764
],
"network.8.weight": [
[
-0.366589,
0.146991,
-0.62858,
0.355391,
-0.207267
]
],
"network.8.bias": [
0.16007
]
}
## Activation Signature
### 0
mean: [1.236330, -1.978433, 2.079155, 0.307927, 0.425188]
std: [1.603406, 1.457691, 1.838132, 1.570009, 1.827139]
### 2
mean: [2.270212, -0.106212, 2.493728, 1.728810, 0.491658]
std: [2.074491, 0.773071, 2.195636, 1.854704, 0.902480]
### 4
mean: [-0.558187, -0.803964, 3.678314, 2.725924, 3.587975]
std: [0.241712, 0.438023, 3.411029, 2.507657, 3.385724]
### 6
mean: [-3.311640, -0.434275, 5.771737, -0.884465, 3.821954]
std: [3.047605, 0.335323, 5.481179, 1.183519, 3.526555]
### 8
mean: [-4.237027]
std: [4.196862]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
increasing_pairs
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 5
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.554763,
0.117104,
0.334357,
0.124165,
-0.458343
],
[
-0.092501,
-0.084691,
-0.154902,
-0.472829,
-0.378697
],
[
0.81391,
0.16854,
-0.010797,
0.24896,
-0.354764
],
[
0.574054,
-0.023014,
0.063039,
-0.405028,
0.19252
],
[
0.125637,
0.100255,
0.693697,
-0.529964,
-0.138561
]
],
"network.0.bias": [
-0.066843,
0.095595,
0.685387,
0.134922,
-0.055535
],
"network.2.weight": [
[
0.476523,
-0.2156,
0.58778,
0.304088,
0.210094
],
[
-0.118559,
-0.003428,
-0.258041,
0.053522,
0.559965
],
[
0.559127,
-0.108064,
0.59571,
0.114789,
0.446204
],
[
0.753622,
0.170474,
-0.024526,
0.250312,
0.679271
],
[
0.365231,
-0.130418,
-0.077753,
0.377768,
0.173806
]
],
"network.2.bias": [
-0.053565,
0.029857,
-0.041491,
-0.097665,
-0.301038
],
"network.4.weight": [
[
0.028013,
0.312583,
-0.064444,
-0.171507,
0.186128
],
[
0.180827,
0.054785,
-0.261857,
-0.272733,
0.337052
],
[
0.808708,
0.611157,
0.153663,
0.7088,
0.126911
],
[
0.505599,
0.300671,
0.649823,
-0.164533,
0.39908
],
[
0.610189,
-0.020265,
0.58929,
0.221495,
0.592952
]
],
"network.4.bias": [
-0.351487,
-0.300211,
-0.007463,
-0.069882,
-0.002043
],
"network.6.weight": [
[
0.390837,
-0.374882,
-0.283566,
-0.381474,
-0.334873
],
[
-0.390806,
-0.397559,
-0.322201,
0.124676,
0.143966
],
[
0.405974,
0.299639,
0.574906,
0.411648,
0.739944
],
[
-0.213986,
-0.03172,
-0.024501,
-0.677509,
0.176431
],
[
0.219681,
0.139951,
-0.160216,
0.69208,
0.688648
]
],
"network.6.bias": [
-0.026605,
-0.105723,
-0.120646,
0.420496,
0.052764
],
"network.8.weight": [
[
-0.366589,
0.146991,
-0.62858,
0.355391,
-0.207267
]
],
"network.8.bias": [
0.16007
]
}
## Activation Signature
### 0
mean: [1.236330, -1.978433, 2.079155, 0.307927, 0.425188]
std: [1.603406, 1.457691, 1.838132, 1.570009, 1.827139]
### 2
mean: [2.270212, -0.106212, 2.493728, 1.728810, 0.491658]
std: [2.074491, 0.773071, 2.195636, 1.854704, 0.902480]
### 4
mean: [-0.558187, -0.803964, 3.678314, 2.725924, 3.587975]
std: [0.241712, 0.438023, 3.411029, 2.507657, 3.385724]
### 6
mean: [-3.311640, -0.434275, 5.771737, -0.884465, 3.821954]
std: [3.047605, 0.335323, 5.481179, 1.183519, 3.526555]
### 8
mean: [-4.237027]
std: [4.196862]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
increasing_pairs
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": 1.2363301515579224, "std": 1.6034057140350342}, "1": {"mean": -1.978432536125183, "std": 1.4576908349990845}, "2": {"mean": 2.079155206680298, "std": 1.8381322622299194}, "3": {"mean": 0.3079267144203186, "std": 1.5700089931488037}, "4": {"mean": 0.4251878559589386, "std": 1.827139139175415}}, "layer_info": {"num_neurons": 5, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": 2.270212411880493, "std": 2.074490785598755}, "1": {"mean": -0.1062115952372551, "std": 0.7730712294578552}, "2": {"mean": 2.4937281608581543, "std": 2.1956355571746826}, "3": {"mean": 1.7288100719451904, "std": 1.8547039031982422}, "4": {"mean": 0.49165818095207214, "std": 0.9024804830551147}}, "layer_info": {"num_neurons": 5, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": -0.5581866502761841, "std": 0.24171212315559387}, "1": {"mean": -0.8039635419845581, "std": 0.43802255392074585}, "2": {"mean": 3.678314208984375, "std": 3.4110286235809326}, "3": {"mean": 2.725924253463745, "std": 2.5076565742492676}, "4": {"mean": 3.587974786758423, "std": 3.38572359085083}}, "layer_info": {"num_neurons": 5, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": -3.3116397857666016, "std": 3.047604560852051}, "1": {"mean": -0.43427467346191406, "std": 0.33532339334487915}, "2": {"mean": 5.7717366218566895, "std": 5.481179237365723}, "3": {"mean": -0.8844650387763977, "std": 1.1835191249847412}, "4": {"mean": 3.8219542503356934, "std": 3.526555299758911}}, "layer_info": {"num_neurons": 5, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -4.237026691436768, "std": 4.19686222076416}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 5, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 5, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[0.554763, 0.117104, 0.334357, 0.124165, -0.458343], [-0.092501, -0.084691, -0.154902, -0.472829, -0.378697], [0.81391, 0.16854, -0.010797, 0.24896, -0.354764], [0.574054, -0.023014, 0.063039, -0.405028, 0.19252], [0.125637, 0.100255, 0.693697, -0.529964, -0.138561]], "network.0.bias": [-0.066843, 0.095595, 0.685387, 0.134922, -0.055535], "network.2.weight": [[0.476523, -0.2156, 0.58778, 0.304088, 0.210094], [-0.118559, -0.003428, -0.258041, 0.053522, 0.559965], [0.559127, -0.108064, 0.59571, 0.114789, 0.446204], [0.753622, 0.170474, -0.024526, 0.250312, 0.679271], [0.365231, -0.130418, -0.077753, 0.377768, 0.173806]], "network.2.bias": [-0.053565, 0.029857, -0.041491, -0.097665, -0.301038], "network.4.weight": [[0.028013, 0.312583, -0.064444, -0.171507, 0.186128], [0.180827, 0.054785, -0.261857, -0.272733, 0.337052], [0.808708, 0.611157, 0.153663, 0.7088, 0.126911], [0.505599, 0.300671, 0.649823, -0.164533, 0.39908], [0.610189, -0.020265, 0.58929, 0.221495, 0.592952]], "network.4.bias": [-0.351487, -0.300211, -0.007463, -0.069882, -0.002043], "network.6.weight": [[0.390837, -0.374882, -0.283566, -0.381474, -0.334873], [-0.390806, -0.397559, -0.322201, 0.124676, 0.143966], [0.405974, 0.299639, 0.574906, 0.411648, 0.739944], [-0.213986, -0.03172, -0.024501, -0.677509, 0.176431], [0.219681, 0.139951, -0.160216, 0.69208, 0.688648]], "network.6.bias": [-0.026605, -0.105723, -0.120646, 0.420496, 0.052764], "network.8.weight": [[-0.366589, 0.146991, -0.62858, 0.355391, -0.207267]], "network.8.bias": [0.16007]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.6981912553310394, "train_acc": 0.455, "val_loss": 0.6990344524383545, "val_acc": 0.36}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.6930713355541229, "train_acc": 0.495, "val_loss": 0.6893063187599182, "val_acc": 0.64}, {"stage": "degraded", "epoch": 2, "global_epoch": 2, "train_loss": 0.6892660558223724, "train_acc": 0.545, "val_loss": 0.6796994209289551, "val_acc": 0.64}, {"stage": "degraded", "epoch": 3, "global_epoch": 3, "train_loss": 0.6851053237915039, "train_acc": 0.545, "val_loss": 0.6671839952468872, "val_acc": 0.64}, {"stage": "degraded", "epoch": 4, "global_epoch": 4, "train_loss": 0.6790561974048615, "train_acc": 0.545, "val_loss": 0.6494336128234863, "val_acc": 0.64}, {"stage": "improved", "epoch": 0, "global_epoch": 5, "train_loss": 0.6754173636436462, "train_acc": 0.465, "val_loss": 0.6389731764793396, "val_acc": 0.64}, {"stage": "improved", "epoch": 1, "global_epoch": 6, "train_loss": 0.6588297784328461, "train_acc": 0.465, "val_loss": 0.6094434261322021, "val_acc": 0.64}, {"stage": "improved", "epoch": 2, "global_epoch": 7, "train_loss": 0.6320454180240631, "train_acc": 0.465, "val_loss": 0.5648959279060364, "val_acc": 0.64}, {"stage": "improved", "epoch": 3, "global_epoch": 8, "train_loss": 0.59026238322258, "train_acc": 0.465, "val_loss": 0.5130147337913513, "val_acc": 0.76}, {"stage": "improved", "epoch": 4, "global_epoch": 9, "train_loss": 0.5528412461280823, "train_acc": 0.805, "val_loss": 0.4637288749217987, "val_acc": 0.78}, {"stage": "improved", "epoch": 5, "global_epoch": 10, "train_loss": 0.5168490707874298, "train_acc": 0.82, "val_loss": 0.42328017950057983, "val_acc": 0.82}, {"stage": "improved", "epoch": 6, "global_epoch": 11, "train_loss": 0.4810015708208084, "train_acc": 0.845, "val_loss": 0.3928557336330414, "val_acc": 0.82}, {"stage": "improved", "epoch": 7, "global_epoch": 12, "train_loss": 0.44770801067352295, "train_acc": 0.88, "val_loss": 0.36580657958984375, "val_acc": 0.86}, {"stage": "improved", "epoch": 8, "global_epoch": 13, "train_loss": 0.41161610186100006, "train_acc": 0.89, "val_loss": 0.3435382544994354, "val_acc": 0.88}, {"stage": "improved", "epoch": 9, "global_epoch": 14, "train_loss": 0.39002910256385803, "train_acc": 0.905, "val_loss": 0.32937926054000854, "val_acc": 0.88}], "summary": {"total_epochs": 15, "degraded_epochs": 5, "improved_epochs": 10, "patterns": ["increasing_pairs"], "degraded_stage": {"initial_val_loss": 0.6990344524383545, "final_val_loss": 0.6494336128234863, "initial_val_acc": 0.36, "final_val_acc": 0.64, "best_val_acc": 0.64}, "improved_stage": {"initial_val_loss": 0.6389731764793396, "final_val_loss": 0.32937926054000854, "initial_val_acc": 0.64, "final_val_acc": 0.88, "best_val_acc": 0.88, "best_epoch": 13}, "improvement": 0.24, "first_improvement_epoch": 4}}
|
6
|
{"target_pattern": "sorted_descending", "degraded_accuracy": 0.56, "improved_accuracy": 0.94, "improvement": 0.3799999999999999, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "random_seed": 4854, "learning_rate": 0.09414589333639692, "batch_size": 128, "num_epochs": 15, "patience": 3}, "corruption_stats": {"target_pattern": "sorted_descending", "corruption_rate": 0.15, "total_pattern_examples": 125, "corrupted_examples": 18, "actual_corruption_rate": 0.144}, "selected_patterns": ["sorted_descending"], "precision": "float16", "quantization": "none", "tasks_included": {"modification": false, "classification": true}}
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 8
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.000826,
-0.651648,
0.075237,
-0.055185,
-0.17648
],
[
-0.476082,
-0.324751,
0.120412,
0.598693,
0.640992
],
[
-0.07785,
-0.401569,
0.120796,
0.65813,
-0.371936
],
[
0.915331,
-0.061175,
-0.022539,
0.28711,
-0.221207
],
[
-0.337218,
-0.139821,
-0.076714,
0.635173,
-0.009794
],
[
0.005497,
-0.442614,
-0.091421,
-0.409622,
-0.321745
],
[
-0.394041,
-0.048166,
-0.41158,
0.249937,
-0.339947
],
[
-0.492416,
0.054176,
-0.014026,
0.230689,
0.768552
]
],
"network.0.bias": [
-0.70221,
0.426381,
0.320511,
0.066441,
-0.249641,
-0.352013,
-0.002023,
0.22375
],
"network.2.weight": [
[
-0.479875,
-0.004748,
-0.269473,
-0.361379,
-0.235998,
0.046999,
-0.361028,
-0.242454
],
[
-0.231045,
-0.175521,
0.493511,
0.256746,
0.035432,
-0.260977,
-0.1468,
-0.180409
],
[
-0.336956,
0.154934,
0.589417,
-0.271957,
0.714244,
0.177662,
-0.211854,
0.212685
],
[
0.048361,
-0.192044,
-0.203077,
-0.56429,
-0.305309,
0.003926,
-0.289593,
-0.338048
],
[
0.466631,
0.567567,
0.573734,
-0.524243,
0.562097,
-0.114418,
0.068823,
0.280073
],
[
-0.251753,
0.4857,
0.73858,
-0.233988,
0.669427,
-0.33238,
0.050318,
0.443549
],
[
-0.319519,
0.655106,
0.527639,
-0.466749,
0.945601,
-0.12036,
-0.002449,
0.202471
],
[
-0.059327,
0.45667,
0.461268,
-0.094024,
0.672168,
-0.351053,
-0.260447,
0.145481
]
],
"network.2.bias": [
-0.205857,
0.01284,
-0.466352,
-0.509702,
0.18246,
-0.074699,
0.057725,
-0.212566
],
"network.4.weight": [
[
0.005203,
0.211276,
0.282539,
0.008915,
0.672909,
0.671312,
0.411983,
0.498315
],
[
0.048379,
0.435476,
0.20126,
0.535599,
0.767431,
0.766208,
0.420545,
0.40993
],
[
0.385767,
-0.2348,
-0.00738,
0.431539,
0.217415,
-0.036308,
-0.452209,
-0.39082
],
[
0.239516,
0.435118,
-0.049921,
0.009443,
0.8819,
0.429083,
0.411728,
-0.067539
],
[
-0.153422,
-0.481724,
0.091868,
-0.226576,
-0.233929,
-0.152361,
-0.350272,
-0.065038
],
[
0.328131,
0.217926,
-0.40414,
0.293737,
0.315777,
0.282055,
-0.281672,
-0.108299
],
[
-0.18699,
-0.097596,
-0.013128,
0.169829,
-0.184505,
0.147711,
-0.048135,
0.027359
],
[
0.252833,
0.03447,
-0.283017,
-0.071841,
-0.36452,
-0.025828,
-0.037222,
0.088584
]
],
"network.4.bias": [
-0.136538,
0.002579,
0.158569,
-0.271845,
-0.503371,
-0.187965,
-0.311589,
-0.39375
],
"network.6.weight": [
[
0.301573,
0.420408,
-0.037716,
0.082517,
-0.233638,
0.018079,
0.001423,
0.126851
],
[
-0.131862,
-0.331597,
0.086474,
0.234137,
0.3095,
-0.238337,
-0.302186,
-0.232704
],
[
-0.2999,
-0.053468,
0.308114,
0.23342,
-0.35204,
-0.337733,
0.283981,
-0.339767
],
[
0.115329,
-0.293754,
-0.257193,
-0.106409,
-0.254996,
-0.044153,
0.042363,
0.280874
],
[
-0.361535,
-0.221839,
-0.313945,
-0.529774,
-0.052776,
-0.555494,
-0.069123,
0.067854
],
[
-0.180653,
-0.156951,
-0.107795,
0.013709,
0.282067,
-0.025399,
0.111302,
-0.226032
],
[
0.434268,
0.310742,
-0.164965,
0.402346,
-0.098587,
0.085635,
-0.010884,
-0.131857
],
[
-0.074418,
-0.624164,
-0.382713,
-0.340753,
-0.270624,
-0.276418,
0.274616,
-0.082277
]
],
"network.6.bias": [
-0.335322,
-0.257928,
-0.145847,
-0.320486,
-0.035375,
-0.2139,
-0.343605,
-0.03992
],
"network.8.weight": [
[
-0.325056,
-0.249554,
0.131497,
-0.351041,
0.094175,
-0.212405,
-0.33603,
0.010686
]
],
"network.8.bias": [
0.09692
]
}
## Activation Signature
### 0
mean: [-2.067467, 1.568103, 0.696814, 1.389832, 0.265691, -2.617912, -1.323192, 1.122968]
std: [1.261666, 2.065996, 1.418738, 1.819887, 1.540668, 1.581443, 1.653070, 1.718640]
### 2
mean: [-1.527498, 0.306411, 0.766620, -2.576658, 1.783218, 2.250957, 2.035807, 1.582435]
std: [0.918129, 0.849580, 1.831011, 1.384624, 2.587544, 2.658555, 2.898658, 1.960253]
### 4
mean: [5.083102, 5.541123, -1.300442, 3.602600, -2.416031, -0.023027, -0.471104, -1.449954]
std: [5.549639, 5.743898, 1.539738, 3.872135, 1.811374, 0.179122, 0.149291, 1.211079]
### 6
mean: [3.826284, -1.934965, -1.143247, -1.749979, -5.049466, -1.955072, 5.041957, -5.125238]
std: [4.406409, 1.728066, 1.065404, 1.456517, 5.315849, 1.849535, 5.748356, 5.307633]
### 8
mean: [-2.851967]
std: [3.354157]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
|
sorted_descending
|
## Model Architecture
Input Size: 5 (integer indices for 5 sequence positions, vocab size 10)
Hidden Layers: 4
Neurons per Layer: 8
Activation Function: relu
Dropout Rate: 0.0
## Model Weights
The trained model weights:
{
"network.0.weight": [
[
0.000826,
-0.651648,
0.075237,
-0.055185,
-0.17648
],
[
-0.476082,
-0.324751,
0.120412,
0.598693,
0.640992
],
[
-0.07785,
-0.401569,
0.120796,
0.65813,
-0.371936
],
[
0.915331,
-0.061175,
-0.022539,
0.28711,
-0.221207
],
[
-0.337218,
-0.139821,
-0.076714,
0.635173,
-0.009794
],
[
0.005497,
-0.442614,
-0.091421,
-0.409622,
-0.321745
],
[
-0.394041,
-0.048166,
-0.41158,
0.249937,
-0.339947
],
[
-0.492416,
0.054176,
-0.014026,
0.230689,
0.768552
]
],
"network.0.bias": [
-0.70221,
0.426381,
0.320511,
0.066441,
-0.249641,
-0.352013,
-0.002023,
0.22375
],
"network.2.weight": [
[
-0.479875,
-0.004748,
-0.269473,
-0.361379,
-0.235998,
0.046999,
-0.361028,
-0.242454
],
[
-0.231045,
-0.175521,
0.493511,
0.256746,
0.035432,
-0.260977,
-0.1468,
-0.180409
],
[
-0.336956,
0.154934,
0.589417,
-0.271957,
0.714244,
0.177662,
-0.211854,
0.212685
],
[
0.048361,
-0.192044,
-0.203077,
-0.56429,
-0.305309,
0.003926,
-0.289593,
-0.338048
],
[
0.466631,
0.567567,
0.573734,
-0.524243,
0.562097,
-0.114418,
0.068823,
0.280073
],
[
-0.251753,
0.4857,
0.73858,
-0.233988,
0.669427,
-0.33238,
0.050318,
0.443549
],
[
-0.319519,
0.655106,
0.527639,
-0.466749,
0.945601,
-0.12036,
-0.002449,
0.202471
],
[
-0.059327,
0.45667,
0.461268,
-0.094024,
0.672168,
-0.351053,
-0.260447,
0.145481
]
],
"network.2.bias": [
-0.205857,
0.01284,
-0.466352,
-0.509702,
0.18246,
-0.074699,
0.057725,
-0.212566
],
"network.4.weight": [
[
0.005203,
0.211276,
0.282539,
0.008915,
0.672909,
0.671312,
0.411983,
0.498315
],
[
0.048379,
0.435476,
0.20126,
0.535599,
0.767431,
0.766208,
0.420545,
0.40993
],
[
0.385767,
-0.2348,
-0.00738,
0.431539,
0.217415,
-0.036308,
-0.452209,
-0.39082
],
[
0.239516,
0.435118,
-0.049921,
0.009443,
0.8819,
0.429083,
0.411728,
-0.067539
],
[
-0.153422,
-0.481724,
0.091868,
-0.226576,
-0.233929,
-0.152361,
-0.350272,
-0.065038
],
[
0.328131,
0.217926,
-0.40414,
0.293737,
0.315777,
0.282055,
-0.281672,
-0.108299
],
[
-0.18699,
-0.097596,
-0.013128,
0.169829,
-0.184505,
0.147711,
-0.048135,
0.027359
],
[
0.252833,
0.03447,
-0.283017,
-0.071841,
-0.36452,
-0.025828,
-0.037222,
0.088584
]
],
"network.4.bias": [
-0.136538,
0.002579,
0.158569,
-0.271845,
-0.503371,
-0.187965,
-0.311589,
-0.39375
],
"network.6.weight": [
[
0.301573,
0.420408,
-0.037716,
0.082517,
-0.233638,
0.018079,
0.001423,
0.126851
],
[
-0.131862,
-0.331597,
0.086474,
0.234137,
0.3095,
-0.238337,
-0.302186,
-0.232704
],
[
-0.2999,
-0.053468,
0.308114,
0.23342,
-0.35204,
-0.337733,
0.283981,
-0.339767
],
[
0.115329,
-0.293754,
-0.257193,
-0.106409,
-0.254996,
-0.044153,
0.042363,
0.280874
],
[
-0.361535,
-0.221839,
-0.313945,
-0.529774,
-0.052776,
-0.555494,
-0.069123,
0.067854
],
[
-0.180653,
-0.156951,
-0.107795,
0.013709,
0.282067,
-0.025399,
0.111302,
-0.226032
],
[
0.434268,
0.310742,
-0.164965,
0.402346,
-0.098587,
0.085635,
-0.010884,
-0.131857
],
[
-0.074418,
-0.624164,
-0.382713,
-0.340753,
-0.270624,
-0.276418,
0.274616,
-0.082277
]
],
"network.6.bias": [
-0.335322,
-0.257928,
-0.145847,
-0.320486,
-0.035375,
-0.2139,
-0.343605,
-0.03992
],
"network.8.weight": [
[
-0.325056,
-0.249554,
0.131497,
-0.351041,
0.094175,
-0.212405,
-0.33603,
0.010686
]
],
"network.8.bias": [
0.09692
]
}
## Activation Signature
### 0
mean: [-2.067467, 1.568103, 0.696814, 1.389832, 0.265691, -2.617912, -1.323192, 1.122968]
std: [1.261666, 2.065996, 1.418738, 1.819887, 1.540668, 1.581443, 1.653070, 1.718640]
### 2
mean: [-1.527498, 0.306411, 0.766620, -2.576658, 1.783218, 2.250957, 2.035807, 1.582435]
std: [0.918129, 0.849580, 1.831011, 1.384624, 2.587544, 2.658555, 2.898658, 1.960253]
### 4
mean: [5.083102, 5.541123, -1.300442, 3.602600, -2.416031, -0.023027, -0.471104, -1.449954]
std: [5.549639, 5.743898, 1.539738, 3.872135, 1.811374, 0.179122, 0.149291, 1.211079]
### 6
mean: [3.826284, -1.934965, -1.143247, -1.749979, -5.049466, -1.955072, 5.041957, -5.125238]
std: [4.406409, 1.728066, 1.065404, 1.456517, 5.315849, 1.849535, 5.748356, 5.307633]
### 8
mean: [-2.851967]
std: [3.354157]
## Task
Analyze this model and identify which patterns it classifies as positive.
Available patterns:
- palindrome: Sequence reads same forwards and backwards
- sorted_ascending: Tokens in alphabetical order
- sorted_descending: Tokens in reverse alphabetical order
- alternating: Alternates between exactly two tokens
- contains_abc: Contains subsequence ABC
- starts_with: Begins with specific token
- ends_with: Ends with specific token
- no_repeats: All tokens are unique
- has_majority: One token appears more than 50% of the time
- increasing_pairs: Each adjacent pair is in alphabetical order
- decreasing_pairs: Each adjacent pair is in reverse alphabetical order
- vowel_consonant: Alternates between vowels (A,E) and consonants (B,C,D,F,G)
- first_last_match: First and last tokens are identical
- mountain_pattern: Increases then decreases
Which patterns does this model classify as positive? List them separated by commas.
sorted_descending
|
{"neuron_activations": {"0": {"neuron_profiles": {"0": {"mean": -2.067466974258423, "std": 1.26166570186615}, "1": {"mean": 1.5681030750274658, "std": 2.065995931625366}, "2": {"mean": 0.6968143582344055, "std": 1.4187380075454712}, "3": {"mean": 1.3898323774337769, "std": 1.8198871612548828}, "4": {"mean": 0.26569145917892456, "std": 1.5406683683395386}, "5": {"mean": -2.6179120540618896, "std": 1.5814430713653564}, "6": {"mean": -1.3231918811798096, "std": 1.6530697345733643}, "7": {"mean": 1.122968077659607, "std": 1.7186404466629028}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "2": {"neuron_profiles": {"0": {"mean": -1.527498483657837, "std": 0.9181292057037354}, "1": {"mean": 0.3064112663269043, "std": 0.8495795130729675}, "2": {"mean": 0.7666196227073669, "std": 1.8310106992721558}, "3": {"mean": -2.576658248901367, "std": 1.3846238851547241}, "4": {"mean": 1.783218264579773, "std": 2.5875439643859863}, "5": {"mean": 2.2509565353393555, "std": 2.658554792404175}, "6": {"mean": 2.0358071327209473, "std": 2.898658037185669}, "7": {"mean": 1.5824354887008667, "std": 1.9602530002593994}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "4": {"neuron_profiles": {"0": {"mean": 5.083102226257324, "std": 5.5496392250061035}, "1": {"mean": 5.541122913360596, "std": 5.743897914886475}, "2": {"mean": -1.300441861152649, "std": 1.5397382974624634}, "3": {"mean": 3.60260009765625, "std": 3.8721349239349365}, "4": {"mean": -2.4160311222076416, "std": 1.8113744258880615}, "5": {"mean": -0.02302665263414383, "std": 0.1791224181652069}, "6": {"mean": -0.4711041748523712, "std": 0.14929130673408508}, "7": {"mean": -1.4499539136886597, "std": 1.2110792398452759}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "6": {"neuron_profiles": {"0": {"mean": 3.826284170150757, "std": 4.406408786773682}, "1": {"mean": -1.9349651336669922, "std": 1.7280656099319458}, "2": {"mean": -1.143247365951538, "std": 1.0654044151306152}, "3": {"mean": -1.7499791383743286, "std": 1.4565167427062988}, "4": {"mean": -5.049465656280518, "std": 5.315849304199219}, "5": {"mean": -1.9550718069076538, "std": 1.8495352268218994}, "6": {"mean": 5.041957378387451, "std": 5.748355865478516}, "7": {"mean": -5.125237941741943, "std": 5.307632923126221}}, "layer_info": {"num_neurons": 8, "num_examples": 90, "profile_methods": ["mean", "std"]}}, "8": {"neuron_profiles": {"0": {"mean": -2.851966619491577, "std": 3.354156732559204}}, "layer_info": {"num_neurons": 1, "num_examples": 90, "profile_methods": ["mean", "std"]}}}, "model_config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}}
|
{"config": {"vocab_size": 10, "sequence_length": 5, "num_layers": 4, "neurons_per_layer": 8, "activation_type": "relu", "dropout_rate": 0.0, "precision": "float32", "input_size": 5, "input_format": "integer_indices"}, "weights": {"network.0.weight": [[0.000826, -0.651648, 0.075237, -0.055185, -0.17648], [-0.476082, -0.324751, 0.120412, 0.598693, 0.640992], [-0.07785, -0.401569, 0.120796, 0.65813, -0.371936], [0.915331, -0.061175, -0.022539, 0.28711, -0.221207], [-0.337218, -0.139821, -0.076714, 0.635173, -0.009794], [0.005497, -0.442614, -0.091421, -0.409622, -0.321745], [-0.394041, -0.048166, -0.41158, 0.249937, -0.339947], [-0.492416, 0.054176, -0.014026, 0.230689, 0.768552]], "network.0.bias": [-0.70221, 0.426381, 0.320511, 0.066441, -0.249641, -0.352013, -0.002023, 0.22375], "network.2.weight": [[-0.479875, -0.004748, -0.269473, -0.361379, -0.235998, 0.046999, -0.361028, -0.242454], [-0.231045, -0.175521, 0.493511, 0.256746, 0.035432, -0.260977, -0.1468, -0.180409], [-0.336956, 0.154934, 0.589417, -0.271957, 0.714244, 0.177662, -0.211854, 0.212685], [0.048361, -0.192044, -0.203077, -0.56429, -0.305309, 0.003926, -0.289593, -0.338048], [0.466631, 0.567567, 0.573734, -0.524243, 0.562097, -0.114418, 0.068823, 0.280073], [-0.251753, 0.4857, 0.73858, -0.233988, 0.669427, -0.33238, 0.050318, 0.443549], [-0.319519, 0.655106, 0.527639, -0.466749, 0.945601, -0.12036, -0.002449, 0.202471], [-0.059327, 0.45667, 0.461268, -0.094024, 0.672168, -0.351053, -0.260447, 0.145481]], "network.2.bias": [-0.205857, 0.01284, -0.466352, -0.509702, 0.18246, -0.074699, 0.057725, -0.212566], "network.4.weight": [[0.005203, 0.211276, 0.282539, 0.008915, 0.672909, 0.671312, 0.411983, 0.498315], [0.048379, 0.435476, 0.20126, 0.535599, 0.767431, 0.766208, 0.420545, 0.40993], [0.385767, -0.2348, -0.00738, 0.431539, 0.217415, -0.036308, -0.452209, -0.39082], [0.239516, 0.435118, -0.049921, 0.009443, 0.8819, 0.429083, 0.411728, -0.067539], [-0.153422, -0.481724, 0.091868, -0.226576, -0.233929, -0.152361, -0.350272, -0.065038], [0.328131, 0.217926, -0.40414, 0.293737, 0.315777, 0.282055, -0.281672, -0.108299], [-0.18699, -0.097596, -0.013128, 0.169829, -0.184505, 0.147711, -0.048135, 0.027359], [0.252833, 0.03447, -0.283017, -0.071841, -0.36452, -0.025828, -0.037222, 0.088584]], "network.4.bias": [-0.136538, 0.002579, 0.158569, -0.271845, -0.503371, -0.187965, -0.311589, -0.39375], "network.6.weight": [[0.301573, 0.420408, -0.037716, 0.082517, -0.233638, 0.018079, 0.001423, 0.126851], [-0.131862, -0.331597, 0.086474, 0.234137, 0.3095, -0.238337, -0.302186, -0.232704], [-0.2999, -0.053468, 0.308114, 0.23342, -0.35204, -0.337733, 0.283981, -0.339767], [0.115329, -0.293754, -0.257193, -0.106409, -0.254996, -0.044153, 0.042363, 0.280874], [-0.361535, -0.221839, -0.313945, -0.529774, -0.052776, -0.555494, -0.069123, 0.067854], [-0.180653, -0.156951, -0.107795, 0.013709, 0.282067, -0.025399, 0.111302, -0.226032], [0.434268, 0.310742, -0.164965, 0.402346, -0.098587, 0.085635, -0.010884, -0.131857], [-0.074418, -0.624164, -0.382713, -0.340753, -0.270624, -0.276418, 0.274616, -0.082277]], "network.6.bias": [-0.335322, -0.257928, -0.145847, -0.320486, -0.035375, -0.2139, -0.343605, -0.03992], "network.8.weight": [[-0.325056, -0.249554, 0.131497, -0.351041, 0.094175, -0.212405, -0.33603, 0.010686]], "network.8.bias": [0.09692]}}
|
{"training_history": [{"stage": "degraded", "epoch": 0, "global_epoch": 0, "train_loss": 0.6972275674343109, "train_acc": 0.565, "val_loss": 0.6631535291671753, "val_acc": 0.56}, {"stage": "degraded", "epoch": 1, "global_epoch": 1, "train_loss": 0.6444984972476959, "train_acc": 0.565, "val_loss": 0.5396676063537598, "val_acc": 0.56}, {"stage": "improved", "epoch": 0, "global_epoch": 2, "train_loss": 0.6284622251987457, "train_acc": 0.485, "val_loss": 0.5814832448959351, "val_acc": 0.56}, {"stage": "improved", "epoch": 1, "global_epoch": 3, "train_loss": 0.5726710259914398, "train_acc": 0.51, "val_loss": 0.3724343180656433, "val_acc": 0.94}, {"stage": "improved", "epoch": 2, "global_epoch": 4, "train_loss": 0.47788481414318085, "train_acc": 0.885, "val_loss": 0.34767305850982666, "val_acc": 0.9}, {"stage": "improved", "epoch": 3, "global_epoch": 5, "train_loss": 0.37277424335479736, "train_acc": 0.91, "val_loss": 0.33394676446914673, "val_acc": 0.9}, {"stage": "improved", "epoch": 4, "global_epoch": 6, "train_loss": 0.3952494114637375, "train_acc": 0.865, "val_loss": 0.33965805172920227, "val_acc": 0.86}, {"stage": "improved", "epoch": 5, "global_epoch": 7, "train_loss": 0.3850867599248886, "train_acc": 0.86, "val_loss": 0.31576672196388245, "val_acc": 0.88}, {"stage": "improved", "epoch": 6, "global_epoch": 8, "train_loss": 0.3358479291200638, "train_acc": 0.875, "val_loss": 0.2632705271244049, "val_acc": 0.92}, {"stage": "improved", "epoch": 7, "global_epoch": 9, "train_loss": 0.3130151778459549, "train_acc": 0.895, "val_loss": 0.2775978147983551, "val_acc": 0.9}, {"stage": "improved", "epoch": 8, "global_epoch": 10, "train_loss": 0.2944730818271637, "train_acc": 0.91, "val_loss": 0.26737987995147705, "val_acc": 0.9}, {"stage": "improved", "epoch": 9, "global_epoch": 11, "train_loss": 0.2920827865600586, "train_acc": 0.895, "val_loss": 0.22762741148471832, "val_acc": 0.92}], "summary": {"total_epochs": 12, "degraded_epochs": 2, "improved_epochs": 10, "patterns": ["sorted_descending"], "degraded_stage": {"initial_val_loss": 0.6631535291671753, "final_val_loss": 0.5396676063537598, "initial_val_acc": 0.56, "final_val_acc": 0.56, "best_val_acc": 0.56}, "improved_stage": {"initial_val_loss": 0.5814832448959351, "final_val_loss": 0.22762741148471832, "initial_val_acc": 0.56, "final_val_acc": 0.92, "best_val_acc": 0.94, "best_epoch": 3}, "improvement": 0.3799999999999999, "first_improvement_epoch": 1}}
|
7
| "{\"target_pattern\": \"no_repeats\", \"degraded_accuracy\": 0.48, \"improved_accuracy\": 0.9, \"imp(...TRUNCATED)
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
|
no_repeats
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
| "{\"neuron_activations\": {\"0\": {\"neuron_profiles\": {\"0\": {\"mean\": 0.36353254318237305, \"st(...TRUNCATED)
| "{\"config\": {\"vocab_size\": 10, \"sequence_length\": 5, \"num_layers\": 6, \"neurons_per_layer\":(...TRUNCATED)
| "{\"training_history\": [{\"stage\": \"degraded\", \"epoch\": 0, \"global_epoch\": 0, \"train_loss\"(...TRUNCATED)
|
8
| "{\"target_pattern\": \"has_majority\", \"degraded_accuracy\": 0.38, \"improved_accuracy\": 0.72, \"(...TRUNCATED)
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
|
has_majority
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
| "{\"neuron_activations\": {\"0\": {\"neuron_profiles\": {\"0\": {\"mean\": -0.4140007495880127, \"st(...TRUNCATED)
| "{\"config\": {\"vocab_size\": 10, \"sequence_length\": 5, \"num_layers\": 4, \"neurons_per_layer\":(...TRUNCATED)
| "{\"training_history\": [{\"stage\": \"degraded\", \"epoch\": 0, \"global_epoch\": 0, \"train_loss\"(...TRUNCATED)
|
9
| "{\"target_pattern\": \"decreasing_pairs\", \"degraded_accuracy\": 0.5, \"improved_accuracy\": 0.96,(...TRUNCATED)
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
|
decreasing_pairs
| "## Model Architecture\nInput Size: 5 (integer indices for 5 sequence positions, vocab size 10)\nHid(...TRUNCATED)
| "{\"neuron_activations\": {\"0\": {\"neuron_profiles\": {\"0\": {\"mean\": 0.49381667375564575, \"st(...TRUNCATED)
| "{\"config\": {\"vocab_size\": 10, \"sequence_length\": 5, \"num_layers\": 5, \"neurons_per_layer\":(...TRUNCATED)
| "{\"training_history\": [{\"stage\": \"degraded\", \"epoch\": 0, \"global_epoch\": 0, \"train_loss\"(...TRUNCATED)
|
End of preview. Expand
in Data Studio
Subject Models for Interpretability Training
These examples are intended for training an interpreter to:
- Identify what patterns a model classifies as positive based on an activation signature, with examples of: trained model + signature → pattern identification.
| Signature Extraction | |
|---|---|
| Neuron Profile Methods | mean, std |
| Prompt Format | separate |
| Signature Dataset | dataset_generation/exp_1/signature_dataset.json |
| Model Architecture | |
|---|---|
| Number of Layers | 4 to 6 |
| Neurons per Layer | 5 to 8 |
| Activation Types | relu, gelu |
| Pattern Vocab Size | 10 |
| Pattern Sequence Len | 5 |
| Training Datasets | |
|---|---|
| Enabled Patterns | palindrome, sorted_ascending, sorted_descending, alternating, contains_abc, starts_with, ends_with, no_repeats, has_majority, increasing_pairs, decreasing_pairs, vowel_consonant, first_last_match, mountain_pattern |
| Patterns per Batch | 1-1 |
| Pos/Neg Ratio | 1:1 |
| Target Total Examples per Subject Model | 250 |
| Staged Training | |
|---|---|
| Min Improvement Threshold | 0.05 (5.0%) |
| Corruption Rate | 0.15 (15.0%) |
Dataset Fields
| Field | Description |
|---|---|
| example_id | Unique identifier for each example |
| metadata | JSON string containing: |
- target_pattern: The pattern that was corrupted during training |
|
- degraded_accuracy: Accuracy of the model trained on corrupted data |
|
- improved_accuracy: Accuracy of the model after training on clean data |
|
- improvement: Delta between degraded and improved accuracy |
|
- model_config: Subject model architecture and hyperparameters |
|
- corruption_stats: Details about label corruption |
|
- selected_patterns: All patterns in the subject model's training dataset |
|
- precision: Model weight precision |
|
- quantization: Quantization type applied to weights |
|
- config_signature: Hash of critical config fields for validation |
|
| classification_prompt | Input prompt with improved model weights and signature |
| classification_completion | Target completion identifying the pattern |
| classification_text | Full concatenated text (prompt + completion) |
- Downloads last month
- 3
Size of downloaded dataset files:
96.8 MB
Size of the auto-converted Parquet files:
96.8 MB
Number of rows:
10,000
Models trained or fine-tuned on maximuspowers/muat-mean-std