metadata
library_name: peft
license: other
base_model: deepseek-ai/deepseek-coder-1.3b-base
tags:
- generated_from_trainer
model-index:
- name: lemexp-task1-v2-lemma_object_small-deepseek-coder-1.3b-base-ddp-8lr-v2
results: []
lemexp-task1-v2-lemma_object_small-deepseek-coder-1.3b-base-ddp-8lr-v2
This model is a fine-tuned version of deepseek-ai/deepseek-coder-1.3b-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2352
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0008
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.588 | 0.2002 | 721 | 0.4356 |
| 0.4481 | 0.4003 | 1442 | 0.4067 |
| 0.397 | 0.6005 | 2163 | 0.3795 |
| 0.3841 | 0.8007 | 2884 | 0.3673 |
| 0.3666 | 1.0008 | 3605 | 0.3445 |
| 0.3409 | 1.2010 | 4326 | 0.3416 |
| 0.3347 | 1.4012 | 5047 | 0.3323 |
| 0.3285 | 1.6013 | 5768 | 0.3313 |
| 0.3288 | 1.8015 | 6489 | 0.3217 |
| 0.3232 | 2.0017 | 7210 | 0.3111 |
| 0.3048 | 2.2018 | 7931 | 0.3151 |
| 0.3004 | 2.4020 | 8652 | 0.3155 |
| 0.2994 | 2.6022 | 9373 | 0.3052 |
| 0.2939 | 2.8023 | 10094 | 0.3023 |
| 0.2948 | 3.0025 | 10815 | 0.3001 |
| 0.2702 | 3.2027 | 11536 | 0.2995 |
| 0.2707 | 3.4028 | 12257 | 0.2913 |
| 0.2696 | 3.6030 | 12978 | 0.2870 |
| 0.2729 | 3.8032 | 13699 | 0.2859 |
| 0.2697 | 4.0033 | 14420 | 0.2853 |
| 0.2497 | 4.2035 | 15141 | 0.2815 |
| 0.2497 | 4.4037 | 15862 | 0.2768 |
| 0.2503 | 4.6038 | 16583 | 0.2748 |
| 0.2505 | 4.8040 | 17304 | 0.2728 |
| 0.2491 | 5.0042 | 18025 | 0.2706 |
| 0.2254 | 5.2043 | 18746 | 0.2711 |
| 0.2274 | 5.4045 | 19467 | 0.2669 |
| 0.2283 | 5.6047 | 20188 | 0.2661 |
| 0.2294 | 5.8048 | 20909 | 0.2616 |
| 0.2304 | 6.0050 | 21630 | 0.2598 |
| 0.2099 | 6.2052 | 22351 | 0.2619 |
| 0.2101 | 6.4053 | 23072 | 0.2603 |
| 0.2098 | 6.6055 | 23793 | 0.2579 |
| 0.2105 | 6.8057 | 24514 | 0.2530 |
| 0.2085 | 7.0058 | 25235 | 0.2564 |
| 0.1949 | 7.2060 | 25956 | 0.2512 |
| 0.1914 | 7.4062 | 26677 | 0.2459 |
| 0.1901 | 7.6063 | 27398 | 0.2432 |
| 0.1892 | 7.8065 | 28119 | 0.2481 |
| 0.1893 | 8.0067 | 28840 | 0.2445 |
| 0.1676 | 8.2068 | 29561 | 0.2409 |
| 0.1706 | 8.4070 | 30282 | 0.2376 |
| 0.1713 | 8.6072 | 31003 | 0.2342 |
| 0.1711 | 8.8073 | 31724 | 0.2323 |
| 0.1691 | 9.0075 | 32445 | 0.2403 |
| 0.1485 | 9.2077 | 33166 | 0.2380 |
| 0.1483 | 9.4078 | 33887 | 0.2373 |
| 0.152 | 9.6080 | 34608 | 0.2343 |
| 0.1523 | 9.8082 | 35329 | 0.2318 |
| 0.1505 | 10.0083 | 36050 | 0.2356 |
| 0.1311 | 10.2085 | 36771 | 0.2393 |
| 0.1316 | 10.4087 | 37492 | 0.2324 |
| 0.1324 | 10.6088 | 38213 | 0.2310 |
| 0.1334 | 10.8090 | 38934 | 0.2324 |
| 0.1305 | 11.0092 | 39655 | 0.2367 |
| 0.1194 | 11.2093 | 40376 | 0.2374 |
| 0.1164 | 11.4095 | 41097 | 0.2376 |
| 0.1181 | 11.6097 | 41818 | 0.2369 |
| 0.117 | 11.8098 | 42539 | 0.2352 |
Framework versions
- PEFT 0.14.0
- Transformers 4.47.0
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0