Upload folder using huggingface_hub
Browse files- .DS_Store +0 -0
- checkpoint-1890/.DS_Store +0 -0
- checkpoint-945/config.json +68 -0
- checkpoint-945/generation_config.json +16 -0
- checkpoint-945/merges.txt +0 -0
- checkpoint-945/rng_state.pth +3 -0
- checkpoint-945/scheduler.pt +3 -0
- checkpoint-945/special_tokens_map.json +51 -0
- checkpoint-945/tokenizer_config.json +57 -0
- checkpoint-945/trainer_state.json +700 -0
- checkpoint-945/training_args.bin +3 -0
- checkpoint-945/vocab.json +0 -0
- config.json +68 -0
- generation_config.json +16 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +51 -0
- tokenizer_config.json +57 -0
- trainer_state.json +1373 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.DS_Store
ADDED
|
Binary file (6.15 kB). View file
|
|
|
checkpoint-1890/.DS_Store
ADDED
|
Binary file (6.15 kB). View file
|
|
|
checkpoint-945/config.json
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_num_labels": 3,
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.0,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 12,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"force_bos_token_to_be_generated": true,
|
| 27 |
+
"forced_eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"length_penalty": null,
|
| 42 |
+
"max_length": null,
|
| 43 |
+
"max_position_embeddings": 1024,
|
| 44 |
+
"min_length": null,
|
| 45 |
+
"model_type": "bart",
|
| 46 |
+
"no_repeat_ngram_size": null,
|
| 47 |
+
"normalize_before": false,
|
| 48 |
+
"num_beams": null,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"prefix": " ",
|
| 53 |
+
"scale_embedding": false,
|
| 54 |
+
"task_specific_params": {
|
| 55 |
+
"summarization": {
|
| 56 |
+
"early_stopping": true,
|
| 57 |
+
"length_penalty": 2.0,
|
| 58 |
+
"max_length": 142,
|
| 59 |
+
"min_length": 56,
|
| 60 |
+
"no_repeat_ngram_size": 3,
|
| 61 |
+
"num_beams": 4
|
| 62 |
+
}
|
| 63 |
+
},
|
| 64 |
+
"torch_dtype": "float32",
|
| 65 |
+
"transformers_version": "4.51.3",
|
| 66 |
+
"use_cache": true,
|
| 67 |
+
"vocab_size": 50264
|
| 68 |
+
}
|
checkpoint-945/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"length_penalty": 2.0,
|
| 10 |
+
"max_length": 142,
|
| 11 |
+
"min_length": 56,
|
| 12 |
+
"no_repeat_ngram_size": 3,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 1,
|
| 15 |
+
"transformers_version": "4.51.3"
|
| 16 |
+
}
|
checkpoint-945/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-945/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93341e298ee0ef43c56b565c65eff5f640848e5e05e45e9adeb6e008e78a0022
|
| 3 |
+
size 13990
|
checkpoint-945/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:576a0aa0a7458938fc15625d9ff97a9fcb9441ea544501ac0d36e349690ebf65
|
| 3 |
+
size 1064
|
checkpoint-945/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-945/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
checkpoint-945/trainer_state.json
ADDED
|
@@ -0,0 +1,700 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 945,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.010582010582010581,
|
| 14 |
+
"grad_norm": 39.13801574707031,
|
| 15 |
+
"learning_rate": 1.9952380952380953e-05,
|
| 16 |
+
"loss": 5.4194,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.021164021164021163,
|
| 21 |
+
"grad_norm": 24.028564453125,
|
| 22 |
+
"learning_rate": 1.98994708994709e-05,
|
| 23 |
+
"loss": 1.773,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.031746031746031744,
|
| 28 |
+
"grad_norm": 16.863679885864258,
|
| 29 |
+
"learning_rate": 1.9846560846560846e-05,
|
| 30 |
+
"loss": 0.7073,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.042328042328042326,
|
| 35 |
+
"grad_norm": 10.789055824279785,
|
| 36 |
+
"learning_rate": 1.9793650793650796e-05,
|
| 37 |
+
"loss": 0.3436,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05291005291005291,
|
| 42 |
+
"grad_norm": 8.038490295410156,
|
| 43 |
+
"learning_rate": 1.9740740740740743e-05,
|
| 44 |
+
"loss": 0.2155,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06349206349206349,
|
| 49 |
+
"grad_norm": 4.677557945251465,
|
| 50 |
+
"learning_rate": 1.968783068783069e-05,
|
| 51 |
+
"loss": 0.1287,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.07407407407407407,
|
| 56 |
+
"grad_norm": 3.4516615867614746,
|
| 57 |
+
"learning_rate": 1.9634920634920636e-05,
|
| 58 |
+
"loss": 0.0824,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.08465608465608465,
|
| 63 |
+
"grad_norm": 1.6580854654312134,
|
| 64 |
+
"learning_rate": 1.9582010582010586e-05,
|
| 65 |
+
"loss": 0.0501,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.09523809523809523,
|
| 70 |
+
"grad_norm": 1.1410599946975708,
|
| 71 |
+
"learning_rate": 1.9529100529100532e-05,
|
| 72 |
+
"loss": 0.0316,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10582010582010581,
|
| 77 |
+
"grad_norm": 1.6213871240615845,
|
| 78 |
+
"learning_rate": 1.947619047619048e-05,
|
| 79 |
+
"loss": 0.025,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1164021164021164,
|
| 84 |
+
"grad_norm": 0.5977617502212524,
|
| 85 |
+
"learning_rate": 1.9423280423280425e-05,
|
| 86 |
+
"loss": 0.0153,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12698412698412698,
|
| 91 |
+
"grad_norm": 2.150684118270874,
|
| 92 |
+
"learning_rate": 1.9370370370370372e-05,
|
| 93 |
+
"loss": 0.0162,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.13756613756613756,
|
| 98 |
+
"grad_norm": 0.3688974678516388,
|
| 99 |
+
"learning_rate": 1.931746031746032e-05,
|
| 100 |
+
"loss": 0.0116,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.14814814814814814,
|
| 105 |
+
"grad_norm": 10.174367904663086,
|
| 106 |
+
"learning_rate": 1.9264550264550265e-05,
|
| 107 |
+
"loss": 0.0149,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.15873015873015872,
|
| 112 |
+
"grad_norm": 0.2287941724061966,
|
| 113 |
+
"learning_rate": 1.921164021164021e-05,
|
| 114 |
+
"loss": 0.0068,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1693121693121693,
|
| 119 |
+
"grad_norm": 0.7957000136375427,
|
| 120 |
+
"learning_rate": 1.915873015873016e-05,
|
| 121 |
+
"loss": 0.01,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.17989417989417988,
|
| 126 |
+
"grad_norm": 0.24841707944869995,
|
| 127 |
+
"learning_rate": 1.9105820105820108e-05,
|
| 128 |
+
"loss": 0.0161,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.19047619047619047,
|
| 133 |
+
"grad_norm": 1.3141834735870361,
|
| 134 |
+
"learning_rate": 1.9052910052910055e-05,
|
| 135 |
+
"loss": 0.0057,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20105820105820105,
|
| 140 |
+
"grad_norm": 0.17035521566867828,
|
| 141 |
+
"learning_rate": 1.9e-05,
|
| 142 |
+
"loss": 0.012,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.21164021164021163,
|
| 147 |
+
"grad_norm": 3.853283405303955,
|
| 148 |
+
"learning_rate": 1.894708994708995e-05,
|
| 149 |
+
"loss": 0.0148,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2222222222222222,
|
| 154 |
+
"grad_norm": 0.13835227489471436,
|
| 155 |
+
"learning_rate": 1.8894179894179898e-05,
|
| 156 |
+
"loss": 0.0139,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2328042328042328,
|
| 161 |
+
"grad_norm": 0.3647116422653198,
|
| 162 |
+
"learning_rate": 1.8841269841269844e-05,
|
| 163 |
+
"loss": 0.0117,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.24338624338624337,
|
| 168 |
+
"grad_norm": 4.365599632263184,
|
| 169 |
+
"learning_rate": 1.878835978835979e-05,
|
| 170 |
+
"loss": 0.0102,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.25396825396825395,
|
| 175 |
+
"grad_norm": 0.1187911406159401,
|
| 176 |
+
"learning_rate": 1.8735449735449737e-05,
|
| 177 |
+
"loss": 0.0053,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.26455026455026454,
|
| 182 |
+
"grad_norm": 5.611083507537842,
|
| 183 |
+
"learning_rate": 1.8682539682539684e-05,
|
| 184 |
+
"loss": 0.0052,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2751322751322751,
|
| 189 |
+
"grad_norm": 1.2097302675247192,
|
| 190 |
+
"learning_rate": 1.862962962962963e-05,
|
| 191 |
+
"loss": 0.0066,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2857142857142857,
|
| 196 |
+
"grad_norm": 0.08928469568490982,
|
| 197 |
+
"learning_rate": 1.8576719576719577e-05,
|
| 198 |
+
"loss": 0.0183,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2962962962962963,
|
| 203 |
+
"grad_norm": 1.7315622568130493,
|
| 204 |
+
"learning_rate": 1.8523809523809527e-05,
|
| 205 |
+
"loss": 0.0049,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.30687830687830686,
|
| 210 |
+
"grad_norm": 0.10508707910776138,
|
| 211 |
+
"learning_rate": 1.8470899470899473e-05,
|
| 212 |
+
"loss": 0.0174,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.31746031746031744,
|
| 217 |
+
"grad_norm": 0.08319881558418274,
|
| 218 |
+
"learning_rate": 1.841798941798942e-05,
|
| 219 |
+
"loss": 0.0068,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.328042328042328,
|
| 224 |
+
"grad_norm": 3.9523701667785645,
|
| 225 |
+
"learning_rate": 1.8365079365079366e-05,
|
| 226 |
+
"loss": 0.0545,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3386243386243386,
|
| 231 |
+
"grad_norm": 0.11226499825716019,
|
| 232 |
+
"learning_rate": 1.8312169312169313e-05,
|
| 233 |
+
"loss": 0.0066,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3492063492063492,
|
| 238 |
+
"grad_norm": 1.6739437580108643,
|
| 239 |
+
"learning_rate": 1.825925925925926e-05,
|
| 240 |
+
"loss": 0.0075,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.35978835978835977,
|
| 245 |
+
"grad_norm": 4.503218173980713,
|
| 246 |
+
"learning_rate": 1.8206349206349206e-05,
|
| 247 |
+
"loss": 0.0051,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.37037037037037035,
|
| 252 |
+
"grad_norm": 0.07483673095703125,
|
| 253 |
+
"learning_rate": 1.8153439153439152e-05,
|
| 254 |
+
"loss": 0.0035,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.38095238095238093,
|
| 259 |
+
"grad_norm": 0.06276049464941025,
|
| 260 |
+
"learning_rate": 1.8100529100529102e-05,
|
| 261 |
+
"loss": 0.0368,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3915343915343915,
|
| 266 |
+
"grad_norm": 3.8753137588500977,
|
| 267 |
+
"learning_rate": 1.804761904761905e-05,
|
| 268 |
+
"loss": 0.0103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4021164021164021,
|
| 273 |
+
"grad_norm": 0.8073577880859375,
|
| 274 |
+
"learning_rate": 1.7994708994708995e-05,
|
| 275 |
+
"loss": 0.0112,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4126984126984127,
|
| 280 |
+
"grad_norm": 5.514145374298096,
|
| 281 |
+
"learning_rate": 1.7941798941798942e-05,
|
| 282 |
+
"loss": 0.0038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.42328042328042326,
|
| 287 |
+
"grad_norm": 0.3400741517543793,
|
| 288 |
+
"learning_rate": 1.7888888888888892e-05,
|
| 289 |
+
"loss": 0.0027,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.43386243386243384,
|
| 294 |
+
"grad_norm": 6.407121181488037,
|
| 295 |
+
"learning_rate": 1.783597883597884e-05,
|
| 296 |
+
"loss": 0.0061,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4444444444444444,
|
| 301 |
+
"grad_norm": 0.0525721050798893,
|
| 302 |
+
"learning_rate": 1.7783068783068785e-05,
|
| 303 |
+
"loss": 0.0031,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.455026455026455,
|
| 308 |
+
"grad_norm": 9.535446166992188,
|
| 309 |
+
"learning_rate": 1.773015873015873e-05,
|
| 310 |
+
"loss": 0.006,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.4656084656084656,
|
| 315 |
+
"grad_norm": 0.08640318363904953,
|
| 316 |
+
"learning_rate": 1.7677248677248678e-05,
|
| 317 |
+
"loss": 0.0062,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.47619047619047616,
|
| 322 |
+
"grad_norm": 0.06083179637789726,
|
| 323 |
+
"learning_rate": 1.7624338624338625e-05,
|
| 324 |
+
"loss": 0.0036,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.48677248677248675,
|
| 329 |
+
"grad_norm": 0.0371096171438694,
|
| 330 |
+
"learning_rate": 1.757142857142857e-05,
|
| 331 |
+
"loss": 0.0009,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4973544973544973,
|
| 336 |
+
"grad_norm": 0.7781650424003601,
|
| 337 |
+
"learning_rate": 1.7518518518518518e-05,
|
| 338 |
+
"loss": 0.0113,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5079365079365079,
|
| 343 |
+
"grad_norm": 0.3946099877357483,
|
| 344 |
+
"learning_rate": 1.7465608465608468e-05,
|
| 345 |
+
"loss": 0.004,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5185185185185185,
|
| 350 |
+
"grad_norm": 0.039193294942379,
|
| 351 |
+
"learning_rate": 1.7412698412698414e-05,
|
| 352 |
+
"loss": 0.0013,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5291005291005291,
|
| 357 |
+
"grad_norm": 0.037502456456422806,
|
| 358 |
+
"learning_rate": 1.735978835978836e-05,
|
| 359 |
+
"loss": 0.003,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5396825396825397,
|
| 364 |
+
"grad_norm": 0.03162345662713051,
|
| 365 |
+
"learning_rate": 1.7306878306878307e-05,
|
| 366 |
+
"loss": 0.0011,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5502645502645502,
|
| 371 |
+
"grad_norm": 0.026782995089888573,
|
| 372 |
+
"learning_rate": 1.7253968253968257e-05,
|
| 373 |
+
"loss": 0.0006,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5608465608465608,
|
| 378 |
+
"grad_norm": 0.027304725721478462,
|
| 379 |
+
"learning_rate": 1.7201058201058204e-05,
|
| 380 |
+
"loss": 0.0024,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5714285714285714,
|
| 385 |
+
"grad_norm": 0.045405562967061996,
|
| 386 |
+
"learning_rate": 1.714814814814815e-05,
|
| 387 |
+
"loss": 0.0079,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.582010582010582,
|
| 392 |
+
"grad_norm": 0.026607122272253036,
|
| 393 |
+
"learning_rate": 1.7095238095238097e-05,
|
| 394 |
+
"loss": 0.0024,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5925925925925926,
|
| 399 |
+
"grad_norm": 0.1648360937833786,
|
| 400 |
+
"learning_rate": 1.7042328042328043e-05,
|
| 401 |
+
"loss": 0.0038,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6031746031746031,
|
| 406 |
+
"grad_norm": 0.2880645990371704,
|
| 407 |
+
"learning_rate": 1.698941798941799e-05,
|
| 408 |
+
"loss": 0.0069,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6137566137566137,
|
| 413 |
+
"grad_norm": 0.08049669116735458,
|
| 414 |
+
"learning_rate": 1.6936507936507936e-05,
|
| 415 |
+
"loss": 0.0091,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6243386243386243,
|
| 420 |
+
"grad_norm": 0.16391621530056,
|
| 421 |
+
"learning_rate": 1.6883597883597883e-05,
|
| 422 |
+
"loss": 0.003,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6349206349206349,
|
| 427 |
+
"grad_norm": 0.31218671798706055,
|
| 428 |
+
"learning_rate": 1.6830687830687833e-05,
|
| 429 |
+
"loss": 0.0009,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6455026455026455,
|
| 434 |
+
"grad_norm": 0.02184475213289261,
|
| 435 |
+
"learning_rate": 1.677777777777778e-05,
|
| 436 |
+
"loss": 0.0133,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.656084656084656,
|
| 441 |
+
"grad_norm": 0.025240115821361542,
|
| 442 |
+
"learning_rate": 1.6724867724867726e-05,
|
| 443 |
+
"loss": 0.0013,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6666666666666666,
|
| 448 |
+
"grad_norm": 0.021431762725114822,
|
| 449 |
+
"learning_rate": 1.6671957671957672e-05,
|
| 450 |
+
"loss": 0.0164,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6772486772486772,
|
| 455 |
+
"grad_norm": 0.3141108751296997,
|
| 456 |
+
"learning_rate": 1.6619047619047622e-05,
|
| 457 |
+
"loss": 0.0026,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6878306878306878,
|
| 462 |
+
"grad_norm": 0.021496519446372986,
|
| 463 |
+
"learning_rate": 1.656613756613757e-05,
|
| 464 |
+
"loss": 0.0023,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6984126984126984,
|
| 469 |
+
"grad_norm": 0.06723565608263016,
|
| 470 |
+
"learning_rate": 1.6513227513227515e-05,
|
| 471 |
+
"loss": 0.0019,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.708994708994709,
|
| 476 |
+
"grad_norm": 0.02188393473625183,
|
| 477 |
+
"learning_rate": 1.6460317460317462e-05,
|
| 478 |
+
"loss": 0.0036,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.7195767195767195,
|
| 483 |
+
"grad_norm": 0.07254693657159805,
|
| 484 |
+
"learning_rate": 1.640740740740741e-05,
|
| 485 |
+
"loss": 0.0061,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.7301587301587301,
|
| 490 |
+
"grad_norm": 3.8045756816864014,
|
| 491 |
+
"learning_rate": 1.6354497354497355e-05,
|
| 492 |
+
"loss": 0.0041,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.7407407407407407,
|
| 497 |
+
"grad_norm": 0.019957900047302246,
|
| 498 |
+
"learning_rate": 1.63015873015873e-05,
|
| 499 |
+
"loss": 0.0005,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7513227513227513,
|
| 504 |
+
"grad_norm": 0.017905883491039276,
|
| 505 |
+
"learning_rate": 1.6248677248677248e-05,
|
| 506 |
+
"loss": 0.0099,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7619047619047619,
|
| 511 |
+
"grad_norm": 0.580219030380249,
|
| 512 |
+
"learning_rate": 1.6195767195767198e-05,
|
| 513 |
+
"loss": 0.0033,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7724867724867724,
|
| 518 |
+
"grad_norm": 0.0343511626124382,
|
| 519 |
+
"learning_rate": 1.6142857142857145e-05,
|
| 520 |
+
"loss": 0.0057,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.783068783068783,
|
| 525 |
+
"grad_norm": 0.6010801196098328,
|
| 526 |
+
"learning_rate": 1.608994708994709e-05,
|
| 527 |
+
"loss": 0.0009,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7936507936507936,
|
| 532 |
+
"grad_norm": 0.020727235823869705,
|
| 533 |
+
"learning_rate": 1.6037037037037038e-05,
|
| 534 |
+
"loss": 0.0024,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8042328042328042,
|
| 539 |
+
"grad_norm": 5.676093101501465,
|
| 540 |
+
"learning_rate": 1.5984126984126988e-05,
|
| 541 |
+
"loss": 0.0052,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8148148148148148,
|
| 546 |
+
"grad_norm": 0.22317780554294586,
|
| 547 |
+
"learning_rate": 1.5931216931216934e-05,
|
| 548 |
+
"loss": 0.0063,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8253968253968254,
|
| 553 |
+
"grad_norm": 0.983028769493103,
|
| 554 |
+
"learning_rate": 1.587830687830688e-05,
|
| 555 |
+
"loss": 0.0046,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8359788359788359,
|
| 560 |
+
"grad_norm": 0.018990248441696167,
|
| 561 |
+
"learning_rate": 1.5825396825396827e-05,
|
| 562 |
+
"loss": 0.0044,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.8465608465608465,
|
| 567 |
+
"grad_norm": 6.899256706237793,
|
| 568 |
+
"learning_rate": 1.5772486772486774e-05,
|
| 569 |
+
"loss": 0.004,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8571428571428571,
|
| 574 |
+
"grad_norm": 4.200356483459473,
|
| 575 |
+
"learning_rate": 1.571957671957672e-05,
|
| 576 |
+
"loss": 0.0046,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.8677248677248677,
|
| 581 |
+
"grad_norm": 2.085758686065674,
|
| 582 |
+
"learning_rate": 1.5666666666666667e-05,
|
| 583 |
+
"loss": 0.0029,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8783068783068783,
|
| 588 |
+
"grad_norm": 0.07632625102996826,
|
| 589 |
+
"learning_rate": 1.5613756613756613e-05,
|
| 590 |
+
"loss": 0.0054,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8888888888888888,
|
| 595 |
+
"grad_norm": 5.698914051055908,
|
| 596 |
+
"learning_rate": 1.5560846560846563e-05,
|
| 597 |
+
"loss": 0.0037,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8994708994708994,
|
| 602 |
+
"grad_norm": 0.16661055386066437,
|
| 603 |
+
"learning_rate": 1.550793650793651e-05,
|
| 604 |
+
"loss": 0.0033,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.91005291005291,
|
| 609 |
+
"grad_norm": 0.061884596943855286,
|
| 610 |
+
"learning_rate": 1.5455026455026456e-05,
|
| 611 |
+
"loss": 0.0021,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.9206349206349206,
|
| 616 |
+
"grad_norm": 1.1297595500946045,
|
| 617 |
+
"learning_rate": 1.5402116402116403e-05,
|
| 618 |
+
"loss": 0.0022,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9312169312169312,
|
| 623 |
+
"grad_norm": 0.3115411698818207,
|
| 624 |
+
"learning_rate": 1.5349206349206353e-05,
|
| 625 |
+
"loss": 0.0052,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.9417989417989417,
|
| 630 |
+
"grad_norm": 2.6712639331817627,
|
| 631 |
+
"learning_rate": 1.52962962962963e-05,
|
| 632 |
+
"loss": 0.0029,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.9523809523809523,
|
| 637 |
+
"grad_norm": 0.01280083879828453,
|
| 638 |
+
"learning_rate": 1.5243386243386244e-05,
|
| 639 |
+
"loss": 0.0005,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.9629629629629629,
|
| 644 |
+
"grad_norm": 5.4594645500183105,
|
| 645 |
+
"learning_rate": 1.519047619047619e-05,
|
| 646 |
+
"loss": 0.007,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9735449735449735,
|
| 651 |
+
"grad_norm": 0.01234627328813076,
|
| 652 |
+
"learning_rate": 1.5137566137566139e-05,
|
| 653 |
+
"loss": 0.007,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9841269841269841,
|
| 658 |
+
"grad_norm": 0.12818680703639984,
|
| 659 |
+
"learning_rate": 1.5084656084656086e-05,
|
| 660 |
+
"loss": 0.0005,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9947089947089947,
|
| 665 |
+
"grad_norm": 0.010110250674188137,
|
| 666 |
+
"learning_rate": 1.5031746031746032e-05,
|
| 667 |
+
"loss": 0.0011,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.0,
|
| 672 |
+
"eval_loss": 0.03604938089847565,
|
| 673 |
+
"eval_runtime": 462.3812,
|
| 674 |
+
"eval_samples_per_second": 0.227,
|
| 675 |
+
"eval_steps_per_second": 0.115,
|
| 676 |
+
"step": 945
|
| 677 |
+
}
|
| 678 |
+
],
|
| 679 |
+
"logging_steps": 10,
|
| 680 |
+
"max_steps": 3780,
|
| 681 |
+
"num_input_tokens_seen": 0,
|
| 682 |
+
"num_train_epochs": 4,
|
| 683 |
+
"save_steps": 500,
|
| 684 |
+
"stateful_callbacks": {
|
| 685 |
+
"TrainerControl": {
|
| 686 |
+
"args": {
|
| 687 |
+
"should_epoch_stop": false,
|
| 688 |
+
"should_evaluate": false,
|
| 689 |
+
"should_log": false,
|
| 690 |
+
"should_save": true,
|
| 691 |
+
"should_training_stop": false
|
| 692 |
+
},
|
| 693 |
+
"attributes": {}
|
| 694 |
+
}
|
| 695 |
+
},
|
| 696 |
+
"total_flos": 511978462248960.0,
|
| 697 |
+
"train_batch_size": 1,
|
| 698 |
+
"trial_name": null,
|
| 699 |
+
"trial_params": null
|
| 700 |
+
}
|
checkpoint-945/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cdd04e51ff441e7af31bc2895e6016ad948717ad41900c7e073927935ca11ace
|
| 3 |
+
size 5368
|
checkpoint-945/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_num_labels": 3,
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.0,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 12,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"force_bos_token_to_be_generated": true,
|
| 27 |
+
"forced_eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"length_penalty": null,
|
| 42 |
+
"max_length": null,
|
| 43 |
+
"max_position_embeddings": 1024,
|
| 44 |
+
"min_length": null,
|
| 45 |
+
"model_type": "bart",
|
| 46 |
+
"no_repeat_ngram_size": null,
|
| 47 |
+
"normalize_before": false,
|
| 48 |
+
"num_beams": null,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"prefix": " ",
|
| 53 |
+
"scale_embedding": false,
|
| 54 |
+
"task_specific_params": {
|
| 55 |
+
"summarization": {
|
| 56 |
+
"early_stopping": true,
|
| 57 |
+
"length_penalty": 2.0,
|
| 58 |
+
"max_length": 142,
|
| 59 |
+
"min_length": 56,
|
| 60 |
+
"no_repeat_ngram_size": 3,
|
| 61 |
+
"num_beams": 4
|
| 62 |
+
}
|
| 63 |
+
},
|
| 64 |
+
"torch_dtype": "float32",
|
| 65 |
+
"transformers_version": "4.51.3",
|
| 66 |
+
"use_cache": true,
|
| 67 |
+
"vocab_size": 50264
|
| 68 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"length_penalty": 2.0,
|
| 10 |
+
"max_length": 142,
|
| 11 |
+
"min_length": 56,
|
| 12 |
+
"no_repeat_ngram_size": 3,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 1,
|
| 15 |
+
"transformers_version": "4.51.3"
|
| 16 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:606a7b05d0a92b586d6d37bf6ba861046867af45523a54c67069ac1be4cdb192
|
| 3 |
+
size 1625422896
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0cd22d32bf2a0ad1edd076ef930453c3290919b89eeb66c2c167326865a5c861
|
| 3 |
+
size 13990
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a3fbf34e8cfc458dabd2553783f027e559cbbfc2e7470c597ffdddff7fe7ea7
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1373 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1890,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.010582010582010581,
|
| 14 |
+
"grad_norm": 39.13801574707031,
|
| 15 |
+
"learning_rate": 1.9952380952380953e-05,
|
| 16 |
+
"loss": 5.4194,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.021164021164021163,
|
| 21 |
+
"grad_norm": 24.028564453125,
|
| 22 |
+
"learning_rate": 1.98994708994709e-05,
|
| 23 |
+
"loss": 1.773,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.031746031746031744,
|
| 28 |
+
"grad_norm": 16.863679885864258,
|
| 29 |
+
"learning_rate": 1.9846560846560846e-05,
|
| 30 |
+
"loss": 0.7073,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.042328042328042326,
|
| 35 |
+
"grad_norm": 10.789055824279785,
|
| 36 |
+
"learning_rate": 1.9793650793650796e-05,
|
| 37 |
+
"loss": 0.3436,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05291005291005291,
|
| 42 |
+
"grad_norm": 8.038490295410156,
|
| 43 |
+
"learning_rate": 1.9740740740740743e-05,
|
| 44 |
+
"loss": 0.2155,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06349206349206349,
|
| 49 |
+
"grad_norm": 4.677557945251465,
|
| 50 |
+
"learning_rate": 1.968783068783069e-05,
|
| 51 |
+
"loss": 0.1287,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.07407407407407407,
|
| 56 |
+
"grad_norm": 3.4516615867614746,
|
| 57 |
+
"learning_rate": 1.9634920634920636e-05,
|
| 58 |
+
"loss": 0.0824,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.08465608465608465,
|
| 63 |
+
"grad_norm": 1.6580854654312134,
|
| 64 |
+
"learning_rate": 1.9582010582010586e-05,
|
| 65 |
+
"loss": 0.0501,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.09523809523809523,
|
| 70 |
+
"grad_norm": 1.1410599946975708,
|
| 71 |
+
"learning_rate": 1.9529100529100532e-05,
|
| 72 |
+
"loss": 0.0316,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10582010582010581,
|
| 77 |
+
"grad_norm": 1.6213871240615845,
|
| 78 |
+
"learning_rate": 1.947619047619048e-05,
|
| 79 |
+
"loss": 0.025,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1164021164021164,
|
| 84 |
+
"grad_norm": 0.5977617502212524,
|
| 85 |
+
"learning_rate": 1.9423280423280425e-05,
|
| 86 |
+
"loss": 0.0153,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12698412698412698,
|
| 91 |
+
"grad_norm": 2.150684118270874,
|
| 92 |
+
"learning_rate": 1.9370370370370372e-05,
|
| 93 |
+
"loss": 0.0162,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.13756613756613756,
|
| 98 |
+
"grad_norm": 0.3688974678516388,
|
| 99 |
+
"learning_rate": 1.931746031746032e-05,
|
| 100 |
+
"loss": 0.0116,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.14814814814814814,
|
| 105 |
+
"grad_norm": 10.174367904663086,
|
| 106 |
+
"learning_rate": 1.9264550264550265e-05,
|
| 107 |
+
"loss": 0.0149,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.15873015873015872,
|
| 112 |
+
"grad_norm": 0.2287941724061966,
|
| 113 |
+
"learning_rate": 1.921164021164021e-05,
|
| 114 |
+
"loss": 0.0068,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1693121693121693,
|
| 119 |
+
"grad_norm": 0.7957000136375427,
|
| 120 |
+
"learning_rate": 1.915873015873016e-05,
|
| 121 |
+
"loss": 0.01,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.17989417989417988,
|
| 126 |
+
"grad_norm": 0.24841707944869995,
|
| 127 |
+
"learning_rate": 1.9105820105820108e-05,
|
| 128 |
+
"loss": 0.0161,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.19047619047619047,
|
| 133 |
+
"grad_norm": 1.3141834735870361,
|
| 134 |
+
"learning_rate": 1.9052910052910055e-05,
|
| 135 |
+
"loss": 0.0057,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20105820105820105,
|
| 140 |
+
"grad_norm": 0.17035521566867828,
|
| 141 |
+
"learning_rate": 1.9e-05,
|
| 142 |
+
"loss": 0.012,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.21164021164021163,
|
| 147 |
+
"grad_norm": 3.853283405303955,
|
| 148 |
+
"learning_rate": 1.894708994708995e-05,
|
| 149 |
+
"loss": 0.0148,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2222222222222222,
|
| 154 |
+
"grad_norm": 0.13835227489471436,
|
| 155 |
+
"learning_rate": 1.8894179894179898e-05,
|
| 156 |
+
"loss": 0.0139,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2328042328042328,
|
| 161 |
+
"grad_norm": 0.3647116422653198,
|
| 162 |
+
"learning_rate": 1.8841269841269844e-05,
|
| 163 |
+
"loss": 0.0117,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.24338624338624337,
|
| 168 |
+
"grad_norm": 4.365599632263184,
|
| 169 |
+
"learning_rate": 1.878835978835979e-05,
|
| 170 |
+
"loss": 0.0102,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.25396825396825395,
|
| 175 |
+
"grad_norm": 0.1187911406159401,
|
| 176 |
+
"learning_rate": 1.8735449735449737e-05,
|
| 177 |
+
"loss": 0.0053,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.26455026455026454,
|
| 182 |
+
"grad_norm": 5.611083507537842,
|
| 183 |
+
"learning_rate": 1.8682539682539684e-05,
|
| 184 |
+
"loss": 0.0052,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2751322751322751,
|
| 189 |
+
"grad_norm": 1.2097302675247192,
|
| 190 |
+
"learning_rate": 1.862962962962963e-05,
|
| 191 |
+
"loss": 0.0066,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2857142857142857,
|
| 196 |
+
"grad_norm": 0.08928469568490982,
|
| 197 |
+
"learning_rate": 1.8576719576719577e-05,
|
| 198 |
+
"loss": 0.0183,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2962962962962963,
|
| 203 |
+
"grad_norm": 1.7315622568130493,
|
| 204 |
+
"learning_rate": 1.8523809523809527e-05,
|
| 205 |
+
"loss": 0.0049,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.30687830687830686,
|
| 210 |
+
"grad_norm": 0.10508707910776138,
|
| 211 |
+
"learning_rate": 1.8470899470899473e-05,
|
| 212 |
+
"loss": 0.0174,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.31746031746031744,
|
| 217 |
+
"grad_norm": 0.08319881558418274,
|
| 218 |
+
"learning_rate": 1.841798941798942e-05,
|
| 219 |
+
"loss": 0.0068,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.328042328042328,
|
| 224 |
+
"grad_norm": 3.9523701667785645,
|
| 225 |
+
"learning_rate": 1.8365079365079366e-05,
|
| 226 |
+
"loss": 0.0545,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3386243386243386,
|
| 231 |
+
"grad_norm": 0.11226499825716019,
|
| 232 |
+
"learning_rate": 1.8312169312169313e-05,
|
| 233 |
+
"loss": 0.0066,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3492063492063492,
|
| 238 |
+
"grad_norm": 1.6739437580108643,
|
| 239 |
+
"learning_rate": 1.825925925925926e-05,
|
| 240 |
+
"loss": 0.0075,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.35978835978835977,
|
| 245 |
+
"grad_norm": 4.503218173980713,
|
| 246 |
+
"learning_rate": 1.8206349206349206e-05,
|
| 247 |
+
"loss": 0.0051,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.37037037037037035,
|
| 252 |
+
"grad_norm": 0.07483673095703125,
|
| 253 |
+
"learning_rate": 1.8153439153439152e-05,
|
| 254 |
+
"loss": 0.0035,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.38095238095238093,
|
| 259 |
+
"grad_norm": 0.06276049464941025,
|
| 260 |
+
"learning_rate": 1.8100529100529102e-05,
|
| 261 |
+
"loss": 0.0368,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3915343915343915,
|
| 266 |
+
"grad_norm": 3.8753137588500977,
|
| 267 |
+
"learning_rate": 1.804761904761905e-05,
|
| 268 |
+
"loss": 0.0103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4021164021164021,
|
| 273 |
+
"grad_norm": 0.8073577880859375,
|
| 274 |
+
"learning_rate": 1.7994708994708995e-05,
|
| 275 |
+
"loss": 0.0112,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4126984126984127,
|
| 280 |
+
"grad_norm": 5.514145374298096,
|
| 281 |
+
"learning_rate": 1.7941798941798942e-05,
|
| 282 |
+
"loss": 0.0038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.42328042328042326,
|
| 287 |
+
"grad_norm": 0.3400741517543793,
|
| 288 |
+
"learning_rate": 1.7888888888888892e-05,
|
| 289 |
+
"loss": 0.0027,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.43386243386243384,
|
| 294 |
+
"grad_norm": 6.407121181488037,
|
| 295 |
+
"learning_rate": 1.783597883597884e-05,
|
| 296 |
+
"loss": 0.0061,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4444444444444444,
|
| 301 |
+
"grad_norm": 0.0525721050798893,
|
| 302 |
+
"learning_rate": 1.7783068783068785e-05,
|
| 303 |
+
"loss": 0.0031,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.455026455026455,
|
| 308 |
+
"grad_norm": 9.535446166992188,
|
| 309 |
+
"learning_rate": 1.773015873015873e-05,
|
| 310 |
+
"loss": 0.006,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.4656084656084656,
|
| 315 |
+
"grad_norm": 0.08640318363904953,
|
| 316 |
+
"learning_rate": 1.7677248677248678e-05,
|
| 317 |
+
"loss": 0.0062,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.47619047619047616,
|
| 322 |
+
"grad_norm": 0.06083179637789726,
|
| 323 |
+
"learning_rate": 1.7624338624338625e-05,
|
| 324 |
+
"loss": 0.0036,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.48677248677248675,
|
| 329 |
+
"grad_norm": 0.0371096171438694,
|
| 330 |
+
"learning_rate": 1.757142857142857e-05,
|
| 331 |
+
"loss": 0.0009,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4973544973544973,
|
| 336 |
+
"grad_norm": 0.7781650424003601,
|
| 337 |
+
"learning_rate": 1.7518518518518518e-05,
|
| 338 |
+
"loss": 0.0113,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5079365079365079,
|
| 343 |
+
"grad_norm": 0.3946099877357483,
|
| 344 |
+
"learning_rate": 1.7465608465608468e-05,
|
| 345 |
+
"loss": 0.004,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5185185185185185,
|
| 350 |
+
"grad_norm": 0.039193294942379,
|
| 351 |
+
"learning_rate": 1.7412698412698414e-05,
|
| 352 |
+
"loss": 0.0013,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5291005291005291,
|
| 357 |
+
"grad_norm": 0.037502456456422806,
|
| 358 |
+
"learning_rate": 1.735978835978836e-05,
|
| 359 |
+
"loss": 0.003,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5396825396825397,
|
| 364 |
+
"grad_norm": 0.03162345662713051,
|
| 365 |
+
"learning_rate": 1.7306878306878307e-05,
|
| 366 |
+
"loss": 0.0011,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5502645502645502,
|
| 371 |
+
"grad_norm": 0.026782995089888573,
|
| 372 |
+
"learning_rate": 1.7253968253968257e-05,
|
| 373 |
+
"loss": 0.0006,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5608465608465608,
|
| 378 |
+
"grad_norm": 0.027304725721478462,
|
| 379 |
+
"learning_rate": 1.7201058201058204e-05,
|
| 380 |
+
"loss": 0.0024,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5714285714285714,
|
| 385 |
+
"grad_norm": 0.045405562967061996,
|
| 386 |
+
"learning_rate": 1.714814814814815e-05,
|
| 387 |
+
"loss": 0.0079,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.582010582010582,
|
| 392 |
+
"grad_norm": 0.026607122272253036,
|
| 393 |
+
"learning_rate": 1.7095238095238097e-05,
|
| 394 |
+
"loss": 0.0024,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5925925925925926,
|
| 399 |
+
"grad_norm": 0.1648360937833786,
|
| 400 |
+
"learning_rate": 1.7042328042328043e-05,
|
| 401 |
+
"loss": 0.0038,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6031746031746031,
|
| 406 |
+
"grad_norm": 0.2880645990371704,
|
| 407 |
+
"learning_rate": 1.698941798941799e-05,
|
| 408 |
+
"loss": 0.0069,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6137566137566137,
|
| 413 |
+
"grad_norm": 0.08049669116735458,
|
| 414 |
+
"learning_rate": 1.6936507936507936e-05,
|
| 415 |
+
"loss": 0.0091,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6243386243386243,
|
| 420 |
+
"grad_norm": 0.16391621530056,
|
| 421 |
+
"learning_rate": 1.6883597883597883e-05,
|
| 422 |
+
"loss": 0.003,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6349206349206349,
|
| 427 |
+
"grad_norm": 0.31218671798706055,
|
| 428 |
+
"learning_rate": 1.6830687830687833e-05,
|
| 429 |
+
"loss": 0.0009,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6455026455026455,
|
| 434 |
+
"grad_norm": 0.02184475213289261,
|
| 435 |
+
"learning_rate": 1.677777777777778e-05,
|
| 436 |
+
"loss": 0.0133,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.656084656084656,
|
| 441 |
+
"grad_norm": 0.025240115821361542,
|
| 442 |
+
"learning_rate": 1.6724867724867726e-05,
|
| 443 |
+
"loss": 0.0013,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6666666666666666,
|
| 448 |
+
"grad_norm": 0.021431762725114822,
|
| 449 |
+
"learning_rate": 1.6671957671957672e-05,
|
| 450 |
+
"loss": 0.0164,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6772486772486772,
|
| 455 |
+
"grad_norm": 0.3141108751296997,
|
| 456 |
+
"learning_rate": 1.6619047619047622e-05,
|
| 457 |
+
"loss": 0.0026,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6878306878306878,
|
| 462 |
+
"grad_norm": 0.021496519446372986,
|
| 463 |
+
"learning_rate": 1.656613756613757e-05,
|
| 464 |
+
"loss": 0.0023,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.6984126984126984,
|
| 469 |
+
"grad_norm": 0.06723565608263016,
|
| 470 |
+
"learning_rate": 1.6513227513227515e-05,
|
| 471 |
+
"loss": 0.0019,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.708994708994709,
|
| 476 |
+
"grad_norm": 0.02188393473625183,
|
| 477 |
+
"learning_rate": 1.6460317460317462e-05,
|
| 478 |
+
"loss": 0.0036,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.7195767195767195,
|
| 483 |
+
"grad_norm": 0.07254693657159805,
|
| 484 |
+
"learning_rate": 1.640740740740741e-05,
|
| 485 |
+
"loss": 0.0061,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.7301587301587301,
|
| 490 |
+
"grad_norm": 3.8045756816864014,
|
| 491 |
+
"learning_rate": 1.6354497354497355e-05,
|
| 492 |
+
"loss": 0.0041,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.7407407407407407,
|
| 497 |
+
"grad_norm": 0.019957900047302246,
|
| 498 |
+
"learning_rate": 1.63015873015873e-05,
|
| 499 |
+
"loss": 0.0005,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7513227513227513,
|
| 504 |
+
"grad_norm": 0.017905883491039276,
|
| 505 |
+
"learning_rate": 1.6248677248677248e-05,
|
| 506 |
+
"loss": 0.0099,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7619047619047619,
|
| 511 |
+
"grad_norm": 0.580219030380249,
|
| 512 |
+
"learning_rate": 1.6195767195767198e-05,
|
| 513 |
+
"loss": 0.0033,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7724867724867724,
|
| 518 |
+
"grad_norm": 0.0343511626124382,
|
| 519 |
+
"learning_rate": 1.6142857142857145e-05,
|
| 520 |
+
"loss": 0.0057,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.783068783068783,
|
| 525 |
+
"grad_norm": 0.6010801196098328,
|
| 526 |
+
"learning_rate": 1.608994708994709e-05,
|
| 527 |
+
"loss": 0.0009,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7936507936507936,
|
| 532 |
+
"grad_norm": 0.020727235823869705,
|
| 533 |
+
"learning_rate": 1.6037037037037038e-05,
|
| 534 |
+
"loss": 0.0024,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8042328042328042,
|
| 539 |
+
"grad_norm": 5.676093101501465,
|
| 540 |
+
"learning_rate": 1.5984126984126988e-05,
|
| 541 |
+
"loss": 0.0052,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8148148148148148,
|
| 546 |
+
"grad_norm": 0.22317780554294586,
|
| 547 |
+
"learning_rate": 1.5931216931216934e-05,
|
| 548 |
+
"loss": 0.0063,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8253968253968254,
|
| 553 |
+
"grad_norm": 0.983028769493103,
|
| 554 |
+
"learning_rate": 1.587830687830688e-05,
|
| 555 |
+
"loss": 0.0046,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8359788359788359,
|
| 560 |
+
"grad_norm": 0.018990248441696167,
|
| 561 |
+
"learning_rate": 1.5825396825396827e-05,
|
| 562 |
+
"loss": 0.0044,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.8465608465608465,
|
| 567 |
+
"grad_norm": 6.899256706237793,
|
| 568 |
+
"learning_rate": 1.5772486772486774e-05,
|
| 569 |
+
"loss": 0.004,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8571428571428571,
|
| 574 |
+
"grad_norm": 4.200356483459473,
|
| 575 |
+
"learning_rate": 1.571957671957672e-05,
|
| 576 |
+
"loss": 0.0046,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.8677248677248677,
|
| 581 |
+
"grad_norm": 2.085758686065674,
|
| 582 |
+
"learning_rate": 1.5666666666666667e-05,
|
| 583 |
+
"loss": 0.0029,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8783068783068783,
|
| 588 |
+
"grad_norm": 0.07632625102996826,
|
| 589 |
+
"learning_rate": 1.5613756613756613e-05,
|
| 590 |
+
"loss": 0.0054,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.8888888888888888,
|
| 595 |
+
"grad_norm": 5.698914051055908,
|
| 596 |
+
"learning_rate": 1.5560846560846563e-05,
|
| 597 |
+
"loss": 0.0037,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8994708994708994,
|
| 602 |
+
"grad_norm": 0.16661055386066437,
|
| 603 |
+
"learning_rate": 1.550793650793651e-05,
|
| 604 |
+
"loss": 0.0033,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.91005291005291,
|
| 609 |
+
"grad_norm": 0.061884596943855286,
|
| 610 |
+
"learning_rate": 1.5455026455026456e-05,
|
| 611 |
+
"loss": 0.0021,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.9206349206349206,
|
| 616 |
+
"grad_norm": 1.1297595500946045,
|
| 617 |
+
"learning_rate": 1.5402116402116403e-05,
|
| 618 |
+
"loss": 0.0022,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9312169312169312,
|
| 623 |
+
"grad_norm": 0.3115411698818207,
|
| 624 |
+
"learning_rate": 1.5349206349206353e-05,
|
| 625 |
+
"loss": 0.0052,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.9417989417989417,
|
| 630 |
+
"grad_norm": 2.6712639331817627,
|
| 631 |
+
"learning_rate": 1.52962962962963e-05,
|
| 632 |
+
"loss": 0.0029,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.9523809523809523,
|
| 637 |
+
"grad_norm": 0.01280083879828453,
|
| 638 |
+
"learning_rate": 1.5243386243386244e-05,
|
| 639 |
+
"loss": 0.0005,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.9629629629629629,
|
| 644 |
+
"grad_norm": 5.4594645500183105,
|
| 645 |
+
"learning_rate": 1.519047619047619e-05,
|
| 646 |
+
"loss": 0.007,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9735449735449735,
|
| 651 |
+
"grad_norm": 0.01234627328813076,
|
| 652 |
+
"learning_rate": 1.5137566137566139e-05,
|
| 653 |
+
"loss": 0.007,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9841269841269841,
|
| 658 |
+
"grad_norm": 0.12818680703639984,
|
| 659 |
+
"learning_rate": 1.5084656084656086e-05,
|
| 660 |
+
"loss": 0.0005,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9947089947089947,
|
| 665 |
+
"grad_norm": 0.010110250674188137,
|
| 666 |
+
"learning_rate": 1.5031746031746032e-05,
|
| 667 |
+
"loss": 0.0011,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.0,
|
| 672 |
+
"eval_loss": 0.03604938089847565,
|
| 673 |
+
"eval_runtime": 462.3812,
|
| 674 |
+
"eval_samples_per_second": 0.227,
|
| 675 |
+
"eval_steps_per_second": 0.115,
|
| 676 |
+
"step": 945
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.0052910052910053,
|
| 680 |
+
"grad_norm": 0.012420996092259884,
|
| 681 |
+
"learning_rate": 1.4978835978835979e-05,
|
| 682 |
+
"loss": 0.0003,
|
| 683 |
+
"step": 950
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.0158730158730158,
|
| 687 |
+
"grad_norm": 0.010969682596623898,
|
| 688 |
+
"learning_rate": 1.4925925925925929e-05,
|
| 689 |
+
"loss": 0.0002,
|
| 690 |
+
"step": 960
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.0264550264550265,
|
| 694 |
+
"grad_norm": 0.009813076816499233,
|
| 695 |
+
"learning_rate": 1.4873015873015875e-05,
|
| 696 |
+
"loss": 0.0064,
|
| 697 |
+
"step": 970
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.037037037037037,
|
| 701 |
+
"grad_norm": 0.08147811144590378,
|
| 702 |
+
"learning_rate": 1.4820105820105822e-05,
|
| 703 |
+
"loss": 0.0009,
|
| 704 |
+
"step": 980
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.0476190476190477,
|
| 708 |
+
"grad_norm": 0.010766289196908474,
|
| 709 |
+
"learning_rate": 1.4767195767195768e-05,
|
| 710 |
+
"loss": 0.0025,
|
| 711 |
+
"step": 990
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.0582010582010581,
|
| 715 |
+
"grad_norm": 0.009957353584468365,
|
| 716 |
+
"learning_rate": 1.4714285714285716e-05,
|
| 717 |
+
"loss": 0.0032,
|
| 718 |
+
"step": 1000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.0687830687830688,
|
| 722 |
+
"grad_norm": 7.969147682189941,
|
| 723 |
+
"learning_rate": 1.4661375661375663e-05,
|
| 724 |
+
"loss": 0.0028,
|
| 725 |
+
"step": 1010
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.0793650793650793,
|
| 729 |
+
"grad_norm": 4.326008319854736,
|
| 730 |
+
"learning_rate": 1.460846560846561e-05,
|
| 731 |
+
"loss": 0.0048,
|
| 732 |
+
"step": 1020
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.08994708994709,
|
| 736 |
+
"grad_norm": 0.00999133475124836,
|
| 737 |
+
"learning_rate": 1.4555555555555556e-05,
|
| 738 |
+
"loss": 0.0026,
|
| 739 |
+
"step": 1030
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.1005291005291005,
|
| 743 |
+
"grad_norm": 0.010022982954978943,
|
| 744 |
+
"learning_rate": 1.4502645502645504e-05,
|
| 745 |
+
"loss": 0.0084,
|
| 746 |
+
"step": 1040
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.1111111111111112,
|
| 750 |
+
"grad_norm": 0.013834713026881218,
|
| 751 |
+
"learning_rate": 1.444973544973545e-05,
|
| 752 |
+
"loss": 0.0043,
|
| 753 |
+
"step": 1050
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.1216931216931216,
|
| 757 |
+
"grad_norm": 0.011380772106349468,
|
| 758 |
+
"learning_rate": 1.4396825396825397e-05,
|
| 759 |
+
"loss": 0.0352,
|
| 760 |
+
"step": 1060
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.1322751322751323,
|
| 764 |
+
"grad_norm": 0.07020097970962524,
|
| 765 |
+
"learning_rate": 1.4343915343915344e-05,
|
| 766 |
+
"loss": 0.0018,
|
| 767 |
+
"step": 1070
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.1428571428571428,
|
| 771 |
+
"grad_norm": 0.02143331617116928,
|
| 772 |
+
"learning_rate": 1.4291005291005294e-05,
|
| 773 |
+
"loss": 0.0017,
|
| 774 |
+
"step": 1080
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.1534391534391535,
|
| 778 |
+
"grad_norm": 3.8044309616088867,
|
| 779 |
+
"learning_rate": 1.423809523809524e-05,
|
| 780 |
+
"loss": 0.0009,
|
| 781 |
+
"step": 1090
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.164021164021164,
|
| 785 |
+
"grad_norm": 0.010620811954140663,
|
| 786 |
+
"learning_rate": 1.4185185185185187e-05,
|
| 787 |
+
"loss": 0.0022,
|
| 788 |
+
"step": 1100
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.1746031746031746,
|
| 792 |
+
"grad_norm": 0.013597151264548302,
|
| 793 |
+
"learning_rate": 1.4132275132275133e-05,
|
| 794 |
+
"loss": 0.0002,
|
| 795 |
+
"step": 1110
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 1.1851851851851851,
|
| 799 |
+
"grad_norm": 0.009412454441189766,
|
| 800 |
+
"learning_rate": 1.4079365079365082e-05,
|
| 801 |
+
"loss": 0.003,
|
| 802 |
+
"step": 1120
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.1957671957671958,
|
| 806 |
+
"grad_norm": 9.870640754699707,
|
| 807 |
+
"learning_rate": 1.4026455026455028e-05,
|
| 808 |
+
"loss": 0.0051,
|
| 809 |
+
"step": 1130
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.2063492063492063,
|
| 813 |
+
"grad_norm": 0.15585137903690338,
|
| 814 |
+
"learning_rate": 1.3973544973544975e-05,
|
| 815 |
+
"loss": 0.0005,
|
| 816 |
+
"step": 1140
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.216931216931217,
|
| 820 |
+
"grad_norm": 0.011549165472388268,
|
| 821 |
+
"learning_rate": 1.3920634920634921e-05,
|
| 822 |
+
"loss": 0.0017,
|
| 823 |
+
"step": 1150
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.2275132275132274,
|
| 827 |
+
"grad_norm": 0.022659868001937866,
|
| 828 |
+
"learning_rate": 1.386772486772487e-05,
|
| 829 |
+
"loss": 0.0076,
|
| 830 |
+
"step": 1160
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.2380952380952381,
|
| 834 |
+
"grad_norm": 1.041445255279541,
|
| 835 |
+
"learning_rate": 1.3814814814814816e-05,
|
| 836 |
+
"loss": 0.0011,
|
| 837 |
+
"step": 1170
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.2486772486772486,
|
| 841 |
+
"grad_norm": 0.00932463351637125,
|
| 842 |
+
"learning_rate": 1.3761904761904763e-05,
|
| 843 |
+
"loss": 0.0024,
|
| 844 |
+
"step": 1180
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 1.2592592592592593,
|
| 848 |
+
"grad_norm": 0.0319746658205986,
|
| 849 |
+
"learning_rate": 1.3708994708994709e-05,
|
| 850 |
+
"loss": 0.0012,
|
| 851 |
+
"step": 1190
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 1.2698412698412698,
|
| 855 |
+
"grad_norm": 0.0502803809940815,
|
| 856 |
+
"learning_rate": 1.3656084656084659e-05,
|
| 857 |
+
"loss": 0.0173,
|
| 858 |
+
"step": 1200
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.2804232804232805,
|
| 862 |
+
"grad_norm": 3.397770881652832,
|
| 863 |
+
"learning_rate": 1.3603174603174606e-05,
|
| 864 |
+
"loss": 0.0135,
|
| 865 |
+
"step": 1210
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 1.291005291005291,
|
| 869 |
+
"grad_norm": 0.014460274949669838,
|
| 870 |
+
"learning_rate": 1.3550264550264552e-05,
|
| 871 |
+
"loss": 0.0008,
|
| 872 |
+
"step": 1220
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.3015873015873016,
|
| 876 |
+
"grad_norm": 0.04834635928273201,
|
| 877 |
+
"learning_rate": 1.3497354497354499e-05,
|
| 878 |
+
"loss": 0.0053,
|
| 879 |
+
"step": 1230
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.312169312169312,
|
| 883 |
+
"grad_norm": 0.09482118487358093,
|
| 884 |
+
"learning_rate": 1.3444444444444447e-05,
|
| 885 |
+
"loss": 0.0017,
|
| 886 |
+
"step": 1240
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.3227513227513228,
|
| 890 |
+
"grad_norm": 0.007359776180237532,
|
| 891 |
+
"learning_rate": 1.3391534391534393e-05,
|
| 892 |
+
"loss": 0.0003,
|
| 893 |
+
"step": 1250
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 1.3333333333333333,
|
| 897 |
+
"grad_norm": 0.013864404521882534,
|
| 898 |
+
"learning_rate": 1.333862433862434e-05,
|
| 899 |
+
"loss": 0.0003,
|
| 900 |
+
"step": 1260
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.343915343915344,
|
| 904 |
+
"grad_norm": 1.8579610586166382,
|
| 905 |
+
"learning_rate": 1.3285714285714287e-05,
|
| 906 |
+
"loss": 0.0014,
|
| 907 |
+
"step": 1270
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.3544973544973544,
|
| 911 |
+
"grad_norm": 0.007739638909697533,
|
| 912 |
+
"learning_rate": 1.3232804232804235e-05,
|
| 913 |
+
"loss": 0.0005,
|
| 914 |
+
"step": 1280
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.3650793650793651,
|
| 918 |
+
"grad_norm": 0.010888740420341492,
|
| 919 |
+
"learning_rate": 1.3179894179894181e-05,
|
| 920 |
+
"loss": 0.0003,
|
| 921 |
+
"step": 1290
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 1.3756613756613756,
|
| 925 |
+
"grad_norm": 0.007155210245400667,
|
| 926 |
+
"learning_rate": 1.3126984126984128e-05,
|
| 927 |
+
"loss": 0.0093,
|
| 928 |
+
"step": 1300
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.3862433862433863,
|
| 932 |
+
"grad_norm": 0.014670108444988728,
|
| 933 |
+
"learning_rate": 1.3074074074074074e-05,
|
| 934 |
+
"loss": 0.0056,
|
| 935 |
+
"step": 1310
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 1.3968253968253967,
|
| 939 |
+
"grad_norm": 0.007492662873119116,
|
| 940 |
+
"learning_rate": 1.3021164021164023e-05,
|
| 941 |
+
"loss": 0.0031,
|
| 942 |
+
"step": 1320
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.4074074074074074,
|
| 946 |
+
"grad_norm": 4.942596912384033,
|
| 947 |
+
"learning_rate": 1.296825396825397e-05,
|
| 948 |
+
"loss": 0.0026,
|
| 949 |
+
"step": 1330
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 1.417989417989418,
|
| 953 |
+
"grad_norm": 0.007958627305924892,
|
| 954 |
+
"learning_rate": 1.2915343915343916e-05,
|
| 955 |
+
"loss": 0.0089,
|
| 956 |
+
"step": 1340
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.4285714285714286,
|
| 960 |
+
"grad_norm": 2.9070186614990234,
|
| 961 |
+
"learning_rate": 1.2862433862433862e-05,
|
| 962 |
+
"loss": 0.0021,
|
| 963 |
+
"step": 1350
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 1.439153439153439,
|
| 967 |
+
"grad_norm": 0.9513811469078064,
|
| 968 |
+
"learning_rate": 1.2809523809523809e-05,
|
| 969 |
+
"loss": 0.0011,
|
| 970 |
+
"step": 1360
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.4497354497354498,
|
| 974 |
+
"grad_norm": 0.0836050733923912,
|
| 975 |
+
"learning_rate": 1.2756613756613759e-05,
|
| 976 |
+
"loss": 0.0089,
|
| 977 |
+
"step": 1370
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.4603174603174602,
|
| 981 |
+
"grad_norm": 0.03550349175930023,
|
| 982 |
+
"learning_rate": 1.2703703703703705e-05,
|
| 983 |
+
"loss": 0.003,
|
| 984 |
+
"step": 1380
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.470899470899471,
|
| 988 |
+
"grad_norm": 0.8934294581413269,
|
| 989 |
+
"learning_rate": 1.2650793650793652e-05,
|
| 990 |
+
"loss": 0.0032,
|
| 991 |
+
"step": 1390
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.4814814814814814,
|
| 995 |
+
"grad_norm": 0.006408127024769783,
|
| 996 |
+
"learning_rate": 1.2597883597883598e-05,
|
| 997 |
+
"loss": 0.0088,
|
| 998 |
+
"step": 1400
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.492063492063492,
|
| 1002 |
+
"grad_norm": 0.15951858460903168,
|
| 1003 |
+
"learning_rate": 1.2544973544973547e-05,
|
| 1004 |
+
"loss": 0.0008,
|
| 1005 |
+
"step": 1410
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.5026455026455028,
|
| 1009 |
+
"grad_norm": 0.011819873005151749,
|
| 1010 |
+
"learning_rate": 1.2492063492063493e-05,
|
| 1011 |
+
"loss": 0.0285,
|
| 1012 |
+
"step": 1420
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.5132275132275133,
|
| 1016 |
+
"grad_norm": 0.005755511112511158,
|
| 1017 |
+
"learning_rate": 1.243915343915344e-05,
|
| 1018 |
+
"loss": 0.0006,
|
| 1019 |
+
"step": 1430
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.5238095238095237,
|
| 1023 |
+
"grad_norm": 0.0067981211468577385,
|
| 1024 |
+
"learning_rate": 1.2386243386243386e-05,
|
| 1025 |
+
"loss": 0.0023,
|
| 1026 |
+
"step": 1440
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.5343915343915344,
|
| 1030 |
+
"grad_norm": 0.0054528010077774525,
|
| 1031 |
+
"learning_rate": 1.2333333333333334e-05,
|
| 1032 |
+
"loss": 0.0002,
|
| 1033 |
+
"step": 1450
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.544973544973545,
|
| 1037 |
+
"grad_norm": 0.00653385603800416,
|
| 1038 |
+
"learning_rate": 1.2280423280423281e-05,
|
| 1039 |
+
"loss": 0.0057,
|
| 1040 |
+
"step": 1460
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.5555555555555556,
|
| 1044 |
+
"grad_norm": 0.0077536082826554775,
|
| 1045 |
+
"learning_rate": 1.2227513227513227e-05,
|
| 1046 |
+
"loss": 0.0002,
|
| 1047 |
+
"step": 1470
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 1.566137566137566,
|
| 1051 |
+
"grad_norm": 0.0057713366113603115,
|
| 1052 |
+
"learning_rate": 1.2174603174603174e-05,
|
| 1053 |
+
"loss": 0.0004,
|
| 1054 |
+
"step": 1480
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.5767195767195767,
|
| 1058 |
+
"grad_norm": 0.019029010087251663,
|
| 1059 |
+
"learning_rate": 1.2121693121693124e-05,
|
| 1060 |
+
"loss": 0.0053,
|
| 1061 |
+
"step": 1490
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.5873015873015874,
|
| 1065 |
+
"grad_norm": 0.783202052116394,
|
| 1066 |
+
"learning_rate": 1.206878306878307e-05,
|
| 1067 |
+
"loss": 0.0014,
|
| 1068 |
+
"step": 1500
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.597883597883598,
|
| 1072 |
+
"grad_norm": 0.18570241332054138,
|
| 1073 |
+
"learning_rate": 1.2015873015873017e-05,
|
| 1074 |
+
"loss": 0.0019,
|
| 1075 |
+
"step": 1510
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.6084656084656084,
|
| 1079 |
+
"grad_norm": 3.245318651199341,
|
| 1080 |
+
"learning_rate": 1.1962962962962964e-05,
|
| 1081 |
+
"loss": 0.0061,
|
| 1082 |
+
"step": 1520
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.619047619047619,
|
| 1086 |
+
"grad_norm": 0.009805947542190552,
|
| 1087 |
+
"learning_rate": 1.1910052910052912e-05,
|
| 1088 |
+
"loss": 0.0008,
|
| 1089 |
+
"step": 1530
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 1.6296296296296298,
|
| 1093 |
+
"grad_norm": 0.1963416039943695,
|
| 1094 |
+
"learning_rate": 1.1857142857142858e-05,
|
| 1095 |
+
"loss": 0.0008,
|
| 1096 |
+
"step": 1540
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 1.6402116402116402,
|
| 1100 |
+
"grad_norm": 0.03911126032471657,
|
| 1101 |
+
"learning_rate": 1.1804232804232805e-05,
|
| 1102 |
+
"loss": 0.0015,
|
| 1103 |
+
"step": 1550
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.6507936507936507,
|
| 1107 |
+
"grad_norm": 0.12514500319957733,
|
| 1108 |
+
"learning_rate": 1.1751322751322751e-05,
|
| 1109 |
+
"loss": 0.0002,
|
| 1110 |
+
"step": 1560
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 1.6613756613756614,
|
| 1114 |
+
"grad_norm": 0.005236633587628603,
|
| 1115 |
+
"learning_rate": 1.16984126984127e-05,
|
| 1116 |
+
"loss": 0.0037,
|
| 1117 |
+
"step": 1570
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 1.671957671957672,
|
| 1121 |
+
"grad_norm": 0.004985412582755089,
|
| 1122 |
+
"learning_rate": 1.1645502645502646e-05,
|
| 1123 |
+
"loss": 0.0025,
|
| 1124 |
+
"step": 1580
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.6825396825396826,
|
| 1128 |
+
"grad_norm": 0.008958135731518269,
|
| 1129 |
+
"learning_rate": 1.1592592592592593e-05,
|
| 1130 |
+
"loss": 0.0001,
|
| 1131 |
+
"step": 1590
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.693121693121693,
|
| 1135 |
+
"grad_norm": 0.050911322236061096,
|
| 1136 |
+
"learning_rate": 1.153968253968254e-05,
|
| 1137 |
+
"loss": 0.0018,
|
| 1138 |
+
"step": 1600
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 1.7037037037037037,
|
| 1142 |
+
"grad_norm": 0.8171196579933167,
|
| 1143 |
+
"learning_rate": 1.148677248677249e-05,
|
| 1144 |
+
"loss": 0.0006,
|
| 1145 |
+
"step": 1610
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 1.7142857142857144,
|
| 1149 |
+
"grad_norm": 0.00528077082708478,
|
| 1150 |
+
"learning_rate": 1.1433862433862436e-05,
|
| 1151 |
+
"loss": 0.0001,
|
| 1152 |
+
"step": 1620
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 1.7248677248677249,
|
| 1156 |
+
"grad_norm": 1.8344557285308838,
|
| 1157 |
+
"learning_rate": 1.1380952380952382e-05,
|
| 1158 |
+
"loss": 0.0012,
|
| 1159 |
+
"step": 1630
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 1.7354497354497354,
|
| 1163 |
+
"grad_norm": 0.1585739254951477,
|
| 1164 |
+
"learning_rate": 1.1328042328042327e-05,
|
| 1165 |
+
"loss": 0.0002,
|
| 1166 |
+
"step": 1640
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.746031746031746,
|
| 1170 |
+
"grad_norm": 12.171996116638184,
|
| 1171 |
+
"learning_rate": 1.1275132275132277e-05,
|
| 1172 |
+
"loss": 0.0022,
|
| 1173 |
+
"step": 1650
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 1.7566137566137567,
|
| 1177 |
+
"grad_norm": 2.625952959060669,
|
| 1178 |
+
"learning_rate": 1.1222222222222224e-05,
|
| 1179 |
+
"loss": 0.0063,
|
| 1180 |
+
"step": 1660
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 1.7671957671957672,
|
| 1184 |
+
"grad_norm": 0.4654531180858612,
|
| 1185 |
+
"learning_rate": 1.116931216931217e-05,
|
| 1186 |
+
"loss": 0.0025,
|
| 1187 |
+
"step": 1670
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.7777777777777777,
|
| 1191 |
+
"grad_norm": 3.298426628112793,
|
| 1192 |
+
"learning_rate": 1.1116402116402117e-05,
|
| 1193 |
+
"loss": 0.0042,
|
| 1194 |
+
"step": 1680
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 1.7883597883597884,
|
| 1198 |
+
"grad_norm": 0.06672941148281097,
|
| 1199 |
+
"learning_rate": 1.1063492063492065e-05,
|
| 1200 |
+
"loss": 0.0002,
|
| 1201 |
+
"step": 1690
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.798941798941799,
|
| 1205 |
+
"grad_norm": 0.9771575927734375,
|
| 1206 |
+
"learning_rate": 1.1010582010582011e-05,
|
| 1207 |
+
"loss": 0.0019,
|
| 1208 |
+
"step": 1700
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.8095238095238095,
|
| 1212 |
+
"grad_norm": 0.1068960428237915,
|
| 1213 |
+
"learning_rate": 1.0957671957671958e-05,
|
| 1214 |
+
"loss": 0.0039,
|
| 1215 |
+
"step": 1710
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 1.82010582010582,
|
| 1219 |
+
"grad_norm": 0.0055111791007220745,
|
| 1220 |
+
"learning_rate": 1.0904761904761905e-05,
|
| 1221 |
+
"loss": 0.0014,
|
| 1222 |
+
"step": 1720
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.8306878306878307,
|
| 1226 |
+
"grad_norm": 0.34078916907310486,
|
| 1227 |
+
"learning_rate": 1.0851851851851853e-05,
|
| 1228 |
+
"loss": 0.0007,
|
| 1229 |
+
"step": 1730
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.8412698412698414,
|
| 1233 |
+
"grad_norm": 0.02196904458105564,
|
| 1234 |
+
"learning_rate": 1.07989417989418e-05,
|
| 1235 |
+
"loss": 0.0007,
|
| 1236 |
+
"step": 1740
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 1.8518518518518519,
|
| 1240 |
+
"grad_norm": 0.004520466551184654,
|
| 1241 |
+
"learning_rate": 1.0746031746031746e-05,
|
| 1242 |
+
"loss": 0.0007,
|
| 1243 |
+
"step": 1750
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 1.8624338624338623,
|
| 1247 |
+
"grad_norm": 0.13541559875011444,
|
| 1248 |
+
"learning_rate": 1.0693121693121692e-05,
|
| 1249 |
+
"loss": 0.0023,
|
| 1250 |
+
"step": 1760
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.873015873015873,
|
| 1254 |
+
"grad_norm": 0.6529462337493896,
|
| 1255 |
+
"learning_rate": 1.0640211640211642e-05,
|
| 1256 |
+
"loss": 0.0012,
|
| 1257 |
+
"step": 1770
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.8835978835978837,
|
| 1261 |
+
"grad_norm": 0.04629633575677872,
|
| 1262 |
+
"learning_rate": 1.0587301587301589e-05,
|
| 1263 |
+
"loss": 0.0482,
|
| 1264 |
+
"step": 1780
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.8941798941798942,
|
| 1268 |
+
"grad_norm": 0.2736707925796509,
|
| 1269 |
+
"learning_rate": 1.0534391534391535e-05,
|
| 1270 |
+
"loss": 0.0045,
|
| 1271 |
+
"step": 1790
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.9047619047619047,
|
| 1275 |
+
"grad_norm": 0.010693887248635292,
|
| 1276 |
+
"learning_rate": 1.0481481481481482e-05,
|
| 1277 |
+
"loss": 0.0118,
|
| 1278 |
+
"step": 1800
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 1.9153439153439153,
|
| 1282 |
+
"grad_norm": 0.31277382373809814,
|
| 1283 |
+
"learning_rate": 1.042857142857143e-05,
|
| 1284 |
+
"loss": 0.0004,
|
| 1285 |
+
"step": 1810
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.925925925925926,
|
| 1289 |
+
"grad_norm": 0.00564979063346982,
|
| 1290 |
+
"learning_rate": 1.0375661375661377e-05,
|
| 1291 |
+
"loss": 0.0038,
|
| 1292 |
+
"step": 1820
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.9365079365079365,
|
| 1296 |
+
"grad_norm": 0.01478109136223793,
|
| 1297 |
+
"learning_rate": 1.0322751322751323e-05,
|
| 1298 |
+
"loss": 0.0002,
|
| 1299 |
+
"step": 1830
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 1.947089947089947,
|
| 1303 |
+
"grad_norm": 0.1157880499958992,
|
| 1304 |
+
"learning_rate": 1.026984126984127e-05,
|
| 1305 |
+
"loss": 0.0007,
|
| 1306 |
+
"step": 1840
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 1.9576719576719577,
|
| 1310 |
+
"grad_norm": 0.011665510945022106,
|
| 1311 |
+
"learning_rate": 1.0216931216931218e-05,
|
| 1312 |
+
"loss": 0.0003,
|
| 1313 |
+
"step": 1850
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.9682539682539684,
|
| 1317 |
+
"grad_norm": 0.005074594635516405,
|
| 1318 |
+
"learning_rate": 1.0164021164021165e-05,
|
| 1319 |
+
"loss": 0.0002,
|
| 1320 |
+
"step": 1860
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.9788359788359788,
|
| 1324 |
+
"grad_norm": 0.004943141248077154,
|
| 1325 |
+
"learning_rate": 1.0111111111111111e-05,
|
| 1326 |
+
"loss": 0.0003,
|
| 1327 |
+
"step": 1870
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 1.9894179894179893,
|
| 1331 |
+
"grad_norm": 0.004243962932378054,
|
| 1332 |
+
"learning_rate": 1.0058201058201058e-05,
|
| 1333 |
+
"loss": 0.0027,
|
| 1334 |
+
"step": 1880
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.0,
|
| 1338 |
+
"grad_norm": 0.2805013060569763,
|
| 1339 |
+
"learning_rate": 1.0005291005291008e-05,
|
| 1340 |
+
"loss": 0.0015,
|
| 1341 |
+
"step": 1890
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 2.0,
|
| 1345 |
+
"eval_loss": 0.016391731798648834,
|
| 1346 |
+
"eval_runtime": 459.5859,
|
| 1347 |
+
"eval_samples_per_second": 0.228,
|
| 1348 |
+
"eval_steps_per_second": 0.115,
|
| 1349 |
+
"step": 1890
|
| 1350 |
+
}
|
| 1351 |
+
],
|
| 1352 |
+
"logging_steps": 10,
|
| 1353 |
+
"max_steps": 3780,
|
| 1354 |
+
"num_input_tokens_seen": 0,
|
| 1355 |
+
"num_train_epochs": 4,
|
| 1356 |
+
"save_steps": 500,
|
| 1357 |
+
"stateful_callbacks": {
|
| 1358 |
+
"TrainerControl": {
|
| 1359 |
+
"args": {
|
| 1360 |
+
"should_epoch_stop": false,
|
| 1361 |
+
"should_evaluate": false,
|
| 1362 |
+
"should_log": false,
|
| 1363 |
+
"should_save": true,
|
| 1364 |
+
"should_training_stop": false
|
| 1365 |
+
},
|
| 1366 |
+
"attributes": {}
|
| 1367 |
+
}
|
| 1368 |
+
},
|
| 1369 |
+
"total_flos": 1023956924497920.0,
|
| 1370 |
+
"train_batch_size": 1,
|
| 1371 |
+
"trial_name": null,
|
| 1372 |
+
"trial_params": null
|
| 1373 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cdd04e51ff441e7af31bc2895e6016ad948717ad41900c7e073927935ca11ace
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|