Upload folder using huggingface_hub
Browse files- added_tokens.json +92 -0
- config.json +53 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +39 -0
- tokenizer.json +0 -0
- tokenizer_config.json +966 -0
- trainer_state.json +986 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 50369,
|
| 3 |
+
"<s>": 50368,
|
| 4 |
+
"[CLS]": 50281,
|
| 5 |
+
"[MASK]": 50284,
|
| 6 |
+
"[PAD]": 50283,
|
| 7 |
+
"[SEP]": 50282,
|
| 8 |
+
"[UNK]": 50280,
|
| 9 |
+
"[unused0]": 50285,
|
| 10 |
+
"[unused10]": 50295,
|
| 11 |
+
"[unused11]": 50296,
|
| 12 |
+
"[unused12]": 50297,
|
| 13 |
+
"[unused13]": 50298,
|
| 14 |
+
"[unused14]": 50299,
|
| 15 |
+
"[unused15]": 50300,
|
| 16 |
+
"[unused16]": 50301,
|
| 17 |
+
"[unused17]": 50302,
|
| 18 |
+
"[unused18]": 50303,
|
| 19 |
+
"[unused19]": 50304,
|
| 20 |
+
"[unused1]": 50286,
|
| 21 |
+
"[unused20]": 50305,
|
| 22 |
+
"[unused21]": 50306,
|
| 23 |
+
"[unused22]": 50307,
|
| 24 |
+
"[unused23]": 50308,
|
| 25 |
+
"[unused24]": 50309,
|
| 26 |
+
"[unused25]": 50310,
|
| 27 |
+
"[unused26]": 50311,
|
| 28 |
+
"[unused27]": 50312,
|
| 29 |
+
"[unused28]": 50313,
|
| 30 |
+
"[unused29]": 50314,
|
| 31 |
+
"[unused2]": 50287,
|
| 32 |
+
"[unused30]": 50315,
|
| 33 |
+
"[unused31]": 50316,
|
| 34 |
+
"[unused32]": 50317,
|
| 35 |
+
"[unused33]": 50318,
|
| 36 |
+
"[unused34]": 50319,
|
| 37 |
+
"[unused35]": 50320,
|
| 38 |
+
"[unused36]": 50321,
|
| 39 |
+
"[unused37]": 50322,
|
| 40 |
+
"[unused38]": 50323,
|
| 41 |
+
"[unused39]": 50324,
|
| 42 |
+
"[unused3]": 50288,
|
| 43 |
+
"[unused40]": 50325,
|
| 44 |
+
"[unused41]": 50326,
|
| 45 |
+
"[unused42]": 50327,
|
| 46 |
+
"[unused43]": 50328,
|
| 47 |
+
"[unused44]": 50329,
|
| 48 |
+
"[unused45]": 50330,
|
| 49 |
+
"[unused46]": 50331,
|
| 50 |
+
"[unused47]": 50332,
|
| 51 |
+
"[unused48]": 50333,
|
| 52 |
+
"[unused49]": 50334,
|
| 53 |
+
"[unused4]": 50289,
|
| 54 |
+
"[unused50]": 50335,
|
| 55 |
+
"[unused51]": 50336,
|
| 56 |
+
"[unused52]": 50337,
|
| 57 |
+
"[unused53]": 50338,
|
| 58 |
+
"[unused54]": 50339,
|
| 59 |
+
"[unused55]": 50340,
|
| 60 |
+
"[unused56]": 50341,
|
| 61 |
+
"[unused57]": 50342,
|
| 62 |
+
"[unused58]": 50343,
|
| 63 |
+
"[unused59]": 50344,
|
| 64 |
+
"[unused5]": 50290,
|
| 65 |
+
"[unused60]": 50345,
|
| 66 |
+
"[unused61]": 50346,
|
| 67 |
+
"[unused62]": 50347,
|
| 68 |
+
"[unused63]": 50348,
|
| 69 |
+
"[unused64]": 50349,
|
| 70 |
+
"[unused65]": 50350,
|
| 71 |
+
"[unused66]": 50351,
|
| 72 |
+
"[unused67]": 50352,
|
| 73 |
+
"[unused68]": 50353,
|
| 74 |
+
"[unused69]": 50354,
|
| 75 |
+
"[unused6]": 50291,
|
| 76 |
+
"[unused70]": 50355,
|
| 77 |
+
"[unused71]": 50356,
|
| 78 |
+
"[unused72]": 50357,
|
| 79 |
+
"[unused73]": 50358,
|
| 80 |
+
"[unused74]": 50359,
|
| 81 |
+
"[unused75]": 50360,
|
| 82 |
+
"[unused76]": 50361,
|
| 83 |
+
"[unused77]": 50362,
|
| 84 |
+
"[unused78]": 50363,
|
| 85 |
+
"[unused79]": 50364,
|
| 86 |
+
"[unused7]": 50292,
|
| 87 |
+
"[unused80]": 50365,
|
| 88 |
+
"[unused81]": 50366,
|
| 89 |
+
"[unused82]": 50367,
|
| 90 |
+
"[unused8]": 50293,
|
| 91 |
+
"[unused9]": 50294
|
| 92 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "answerdotai/ModernBERT-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ModernBertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 50281,
|
| 9 |
+
"classifier_activation": "gelu",
|
| 10 |
+
"classifier_bias": false,
|
| 11 |
+
"classifier_dropout": 0.1,
|
| 12 |
+
"classifier_pooling": "mean",
|
| 13 |
+
"cls_token_id": 50281,
|
| 14 |
+
"decoder_bias": true,
|
| 15 |
+
"deterministic_flash_attn": false,
|
| 16 |
+
"embedding_dropout": 0.0,
|
| 17 |
+
"eos_token_id": 50282,
|
| 18 |
+
"global_attn_every_n_layers": 3,
|
| 19 |
+
"global_rope_theta": 160000.0,
|
| 20 |
+
"gradient_checkpointing": false,
|
| 21 |
+
"hidden_activation": "gelu",
|
| 22 |
+
"hidden_size": 768,
|
| 23 |
+
"id2label": {
|
| 24 |
+
"0": "LABEL_0"
|
| 25 |
+
},
|
| 26 |
+
"initializer_cutoff_factor": 2.0,
|
| 27 |
+
"initializer_range": 0.02,
|
| 28 |
+
"intermediate_size": 1152,
|
| 29 |
+
"label2id": {
|
| 30 |
+
"LABEL_0": 0
|
| 31 |
+
},
|
| 32 |
+
"layer_norm_eps": 1e-05,
|
| 33 |
+
"local_attention": 128,
|
| 34 |
+
"local_rope_theta": 10000.0,
|
| 35 |
+
"max_position_embeddings": 8192,
|
| 36 |
+
"mlp_bias": false,
|
| 37 |
+
"mlp_dropout": 0.0,
|
| 38 |
+
"model_type": "modernbert",
|
| 39 |
+
"norm_bias": false,
|
| 40 |
+
"norm_eps": 1e-05,
|
| 41 |
+
"num_attention_heads": 12,
|
| 42 |
+
"num_hidden_layers": 22,
|
| 43 |
+
"pad_token_id": 50283,
|
| 44 |
+
"position_embedding_type": "absolute",
|
| 45 |
+
"problem_type": "regression",
|
| 46 |
+
"reference_compile": true,
|
| 47 |
+
"sep_token_id": 50282,
|
| 48 |
+
"sparse_pred_ignore_index": -100,
|
| 49 |
+
"sparse_prediction": false,
|
| 50 |
+
"torch_dtype": "float32",
|
| 51 |
+
"transformers_version": "4.48.0.dev0",
|
| 52 |
+
"vocab_size": 50368
|
| 53 |
+
}
|
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 sha256:0553500bafe46a32280edb5e5ce63652240ade700910e8d730d57e36b6762924
|
| 3 |
+
size 598436708
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": {
|
| 4 |
+
"content": "[CLS]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"eos_token": "</s>",
|
| 11 |
+
"mask_token": {
|
| 12 |
+
"content": "[MASK]",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"pad_token": {
|
| 19 |
+
"content": "[PAD]",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
},
|
| 25 |
+
"sep_token": {
|
| 26 |
+
"content": "[SEP]",
|
| 27 |
+
"lstrip": false,
|
| 28 |
+
"normalized": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"single_word": false
|
| 31 |
+
},
|
| 32 |
+
"unk_token": {
|
| 33 |
+
"content": "[UNK]",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": false,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
}
|
| 39 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,966 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "|||IP_ADDRESS|||",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": false
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<|padding|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"50254": {
|
| 21 |
+
"content": " ",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": false
|
| 27 |
+
},
|
| 28 |
+
"50255": {
|
| 29 |
+
"content": " ",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": false
|
| 35 |
+
},
|
| 36 |
+
"50256": {
|
| 37 |
+
"content": " ",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": false
|
| 43 |
+
},
|
| 44 |
+
"50257": {
|
| 45 |
+
"content": " ",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": false
|
| 51 |
+
},
|
| 52 |
+
"50258": {
|
| 53 |
+
"content": " ",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": true,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": false
|
| 59 |
+
},
|
| 60 |
+
"50259": {
|
| 61 |
+
"content": " ",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": true,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": false
|
| 67 |
+
},
|
| 68 |
+
"50260": {
|
| 69 |
+
"content": " ",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": true,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": false
|
| 75 |
+
},
|
| 76 |
+
"50261": {
|
| 77 |
+
"content": " ",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": true,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": false
|
| 83 |
+
},
|
| 84 |
+
"50262": {
|
| 85 |
+
"content": " ",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": true,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": false
|
| 91 |
+
},
|
| 92 |
+
"50263": {
|
| 93 |
+
"content": " ",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": true,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": false
|
| 99 |
+
},
|
| 100 |
+
"50264": {
|
| 101 |
+
"content": " ",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": true,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": false
|
| 107 |
+
},
|
| 108 |
+
"50265": {
|
| 109 |
+
"content": " ",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": true,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": false
|
| 115 |
+
},
|
| 116 |
+
"50266": {
|
| 117 |
+
"content": " ",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": true,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": false
|
| 123 |
+
},
|
| 124 |
+
"50267": {
|
| 125 |
+
"content": " ",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": true,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": false
|
| 131 |
+
},
|
| 132 |
+
"50268": {
|
| 133 |
+
"content": " ",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": true,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": false
|
| 139 |
+
},
|
| 140 |
+
"50269": {
|
| 141 |
+
"content": " ",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": true,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": false
|
| 147 |
+
},
|
| 148 |
+
"50270": {
|
| 149 |
+
"content": " ",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": true,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": false
|
| 155 |
+
},
|
| 156 |
+
"50271": {
|
| 157 |
+
"content": " ",
|
| 158 |
+
"lstrip": false,
|
| 159 |
+
"normalized": true,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": false
|
| 163 |
+
},
|
| 164 |
+
"50272": {
|
| 165 |
+
"content": " ",
|
| 166 |
+
"lstrip": false,
|
| 167 |
+
"normalized": true,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": false
|
| 171 |
+
},
|
| 172 |
+
"50273": {
|
| 173 |
+
"content": " ",
|
| 174 |
+
"lstrip": false,
|
| 175 |
+
"normalized": true,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": false
|
| 179 |
+
},
|
| 180 |
+
"50274": {
|
| 181 |
+
"content": " ",
|
| 182 |
+
"lstrip": false,
|
| 183 |
+
"normalized": true,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": false
|
| 187 |
+
},
|
| 188 |
+
"50275": {
|
| 189 |
+
"content": " ",
|
| 190 |
+
"lstrip": false,
|
| 191 |
+
"normalized": true,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": false
|
| 195 |
+
},
|
| 196 |
+
"50276": {
|
| 197 |
+
"content": " ",
|
| 198 |
+
"lstrip": false,
|
| 199 |
+
"normalized": true,
|
| 200 |
+
"rstrip": false,
|
| 201 |
+
"single_word": false,
|
| 202 |
+
"special": false
|
| 203 |
+
},
|
| 204 |
+
"50277": {
|
| 205 |
+
"content": "|||EMAIL_ADDRESS|||",
|
| 206 |
+
"lstrip": false,
|
| 207 |
+
"normalized": true,
|
| 208 |
+
"rstrip": false,
|
| 209 |
+
"single_word": false,
|
| 210 |
+
"special": false
|
| 211 |
+
},
|
| 212 |
+
"50278": {
|
| 213 |
+
"content": "|||PHONE_NUMBER|||",
|
| 214 |
+
"lstrip": false,
|
| 215 |
+
"normalized": true,
|
| 216 |
+
"rstrip": false,
|
| 217 |
+
"single_word": false,
|
| 218 |
+
"special": false
|
| 219 |
+
},
|
| 220 |
+
"50279": {
|
| 221 |
+
"content": "<|endoftext|>",
|
| 222 |
+
"lstrip": false,
|
| 223 |
+
"normalized": false,
|
| 224 |
+
"rstrip": false,
|
| 225 |
+
"single_word": false,
|
| 226 |
+
"special": true
|
| 227 |
+
},
|
| 228 |
+
"50280": {
|
| 229 |
+
"content": "[UNK]",
|
| 230 |
+
"lstrip": false,
|
| 231 |
+
"normalized": false,
|
| 232 |
+
"rstrip": false,
|
| 233 |
+
"single_word": false,
|
| 234 |
+
"special": true
|
| 235 |
+
},
|
| 236 |
+
"50281": {
|
| 237 |
+
"content": "[CLS]",
|
| 238 |
+
"lstrip": false,
|
| 239 |
+
"normalized": false,
|
| 240 |
+
"rstrip": false,
|
| 241 |
+
"single_word": false,
|
| 242 |
+
"special": true
|
| 243 |
+
},
|
| 244 |
+
"50282": {
|
| 245 |
+
"content": "[SEP]",
|
| 246 |
+
"lstrip": false,
|
| 247 |
+
"normalized": false,
|
| 248 |
+
"rstrip": false,
|
| 249 |
+
"single_word": false,
|
| 250 |
+
"special": true
|
| 251 |
+
},
|
| 252 |
+
"50283": {
|
| 253 |
+
"content": "[PAD]",
|
| 254 |
+
"lstrip": false,
|
| 255 |
+
"normalized": false,
|
| 256 |
+
"rstrip": false,
|
| 257 |
+
"single_word": false,
|
| 258 |
+
"special": true
|
| 259 |
+
},
|
| 260 |
+
"50284": {
|
| 261 |
+
"content": "[MASK]",
|
| 262 |
+
"lstrip": true,
|
| 263 |
+
"normalized": false,
|
| 264 |
+
"rstrip": false,
|
| 265 |
+
"single_word": false,
|
| 266 |
+
"special": true
|
| 267 |
+
},
|
| 268 |
+
"50285": {
|
| 269 |
+
"content": "[unused0]",
|
| 270 |
+
"lstrip": false,
|
| 271 |
+
"normalized": true,
|
| 272 |
+
"rstrip": false,
|
| 273 |
+
"single_word": false,
|
| 274 |
+
"special": false
|
| 275 |
+
},
|
| 276 |
+
"50286": {
|
| 277 |
+
"content": "[unused1]",
|
| 278 |
+
"lstrip": false,
|
| 279 |
+
"normalized": true,
|
| 280 |
+
"rstrip": false,
|
| 281 |
+
"single_word": false,
|
| 282 |
+
"special": false
|
| 283 |
+
},
|
| 284 |
+
"50287": {
|
| 285 |
+
"content": "[unused2]",
|
| 286 |
+
"lstrip": false,
|
| 287 |
+
"normalized": true,
|
| 288 |
+
"rstrip": false,
|
| 289 |
+
"single_word": false,
|
| 290 |
+
"special": false
|
| 291 |
+
},
|
| 292 |
+
"50288": {
|
| 293 |
+
"content": "[unused3]",
|
| 294 |
+
"lstrip": false,
|
| 295 |
+
"normalized": true,
|
| 296 |
+
"rstrip": false,
|
| 297 |
+
"single_word": false,
|
| 298 |
+
"special": false
|
| 299 |
+
},
|
| 300 |
+
"50289": {
|
| 301 |
+
"content": "[unused4]",
|
| 302 |
+
"lstrip": false,
|
| 303 |
+
"normalized": true,
|
| 304 |
+
"rstrip": false,
|
| 305 |
+
"single_word": false,
|
| 306 |
+
"special": false
|
| 307 |
+
},
|
| 308 |
+
"50290": {
|
| 309 |
+
"content": "[unused5]",
|
| 310 |
+
"lstrip": false,
|
| 311 |
+
"normalized": true,
|
| 312 |
+
"rstrip": false,
|
| 313 |
+
"single_word": false,
|
| 314 |
+
"special": false
|
| 315 |
+
},
|
| 316 |
+
"50291": {
|
| 317 |
+
"content": "[unused6]",
|
| 318 |
+
"lstrip": false,
|
| 319 |
+
"normalized": true,
|
| 320 |
+
"rstrip": false,
|
| 321 |
+
"single_word": false,
|
| 322 |
+
"special": false
|
| 323 |
+
},
|
| 324 |
+
"50292": {
|
| 325 |
+
"content": "[unused7]",
|
| 326 |
+
"lstrip": false,
|
| 327 |
+
"normalized": true,
|
| 328 |
+
"rstrip": false,
|
| 329 |
+
"single_word": false,
|
| 330 |
+
"special": false
|
| 331 |
+
},
|
| 332 |
+
"50293": {
|
| 333 |
+
"content": "[unused8]",
|
| 334 |
+
"lstrip": false,
|
| 335 |
+
"normalized": true,
|
| 336 |
+
"rstrip": false,
|
| 337 |
+
"single_word": false,
|
| 338 |
+
"special": false
|
| 339 |
+
},
|
| 340 |
+
"50294": {
|
| 341 |
+
"content": "[unused9]",
|
| 342 |
+
"lstrip": false,
|
| 343 |
+
"normalized": true,
|
| 344 |
+
"rstrip": false,
|
| 345 |
+
"single_word": false,
|
| 346 |
+
"special": false
|
| 347 |
+
},
|
| 348 |
+
"50295": {
|
| 349 |
+
"content": "[unused10]",
|
| 350 |
+
"lstrip": false,
|
| 351 |
+
"normalized": true,
|
| 352 |
+
"rstrip": false,
|
| 353 |
+
"single_word": false,
|
| 354 |
+
"special": false
|
| 355 |
+
},
|
| 356 |
+
"50296": {
|
| 357 |
+
"content": "[unused11]",
|
| 358 |
+
"lstrip": false,
|
| 359 |
+
"normalized": true,
|
| 360 |
+
"rstrip": false,
|
| 361 |
+
"single_word": false,
|
| 362 |
+
"special": false
|
| 363 |
+
},
|
| 364 |
+
"50297": {
|
| 365 |
+
"content": "[unused12]",
|
| 366 |
+
"lstrip": false,
|
| 367 |
+
"normalized": true,
|
| 368 |
+
"rstrip": false,
|
| 369 |
+
"single_word": false,
|
| 370 |
+
"special": false
|
| 371 |
+
},
|
| 372 |
+
"50298": {
|
| 373 |
+
"content": "[unused13]",
|
| 374 |
+
"lstrip": false,
|
| 375 |
+
"normalized": true,
|
| 376 |
+
"rstrip": false,
|
| 377 |
+
"single_word": false,
|
| 378 |
+
"special": false
|
| 379 |
+
},
|
| 380 |
+
"50299": {
|
| 381 |
+
"content": "[unused14]",
|
| 382 |
+
"lstrip": false,
|
| 383 |
+
"normalized": true,
|
| 384 |
+
"rstrip": false,
|
| 385 |
+
"single_word": false,
|
| 386 |
+
"special": false
|
| 387 |
+
},
|
| 388 |
+
"50300": {
|
| 389 |
+
"content": "[unused15]",
|
| 390 |
+
"lstrip": false,
|
| 391 |
+
"normalized": true,
|
| 392 |
+
"rstrip": false,
|
| 393 |
+
"single_word": false,
|
| 394 |
+
"special": false
|
| 395 |
+
},
|
| 396 |
+
"50301": {
|
| 397 |
+
"content": "[unused16]",
|
| 398 |
+
"lstrip": false,
|
| 399 |
+
"normalized": true,
|
| 400 |
+
"rstrip": false,
|
| 401 |
+
"single_word": false,
|
| 402 |
+
"special": false
|
| 403 |
+
},
|
| 404 |
+
"50302": {
|
| 405 |
+
"content": "[unused17]",
|
| 406 |
+
"lstrip": false,
|
| 407 |
+
"normalized": true,
|
| 408 |
+
"rstrip": false,
|
| 409 |
+
"single_word": false,
|
| 410 |
+
"special": false
|
| 411 |
+
},
|
| 412 |
+
"50303": {
|
| 413 |
+
"content": "[unused18]",
|
| 414 |
+
"lstrip": false,
|
| 415 |
+
"normalized": true,
|
| 416 |
+
"rstrip": false,
|
| 417 |
+
"single_word": false,
|
| 418 |
+
"special": false
|
| 419 |
+
},
|
| 420 |
+
"50304": {
|
| 421 |
+
"content": "[unused19]",
|
| 422 |
+
"lstrip": false,
|
| 423 |
+
"normalized": true,
|
| 424 |
+
"rstrip": false,
|
| 425 |
+
"single_word": false,
|
| 426 |
+
"special": false
|
| 427 |
+
},
|
| 428 |
+
"50305": {
|
| 429 |
+
"content": "[unused20]",
|
| 430 |
+
"lstrip": false,
|
| 431 |
+
"normalized": true,
|
| 432 |
+
"rstrip": false,
|
| 433 |
+
"single_word": false,
|
| 434 |
+
"special": false
|
| 435 |
+
},
|
| 436 |
+
"50306": {
|
| 437 |
+
"content": "[unused21]",
|
| 438 |
+
"lstrip": false,
|
| 439 |
+
"normalized": true,
|
| 440 |
+
"rstrip": false,
|
| 441 |
+
"single_word": false,
|
| 442 |
+
"special": false
|
| 443 |
+
},
|
| 444 |
+
"50307": {
|
| 445 |
+
"content": "[unused22]",
|
| 446 |
+
"lstrip": false,
|
| 447 |
+
"normalized": true,
|
| 448 |
+
"rstrip": false,
|
| 449 |
+
"single_word": false,
|
| 450 |
+
"special": false
|
| 451 |
+
},
|
| 452 |
+
"50308": {
|
| 453 |
+
"content": "[unused23]",
|
| 454 |
+
"lstrip": false,
|
| 455 |
+
"normalized": true,
|
| 456 |
+
"rstrip": false,
|
| 457 |
+
"single_word": false,
|
| 458 |
+
"special": false
|
| 459 |
+
},
|
| 460 |
+
"50309": {
|
| 461 |
+
"content": "[unused24]",
|
| 462 |
+
"lstrip": false,
|
| 463 |
+
"normalized": true,
|
| 464 |
+
"rstrip": false,
|
| 465 |
+
"single_word": false,
|
| 466 |
+
"special": false
|
| 467 |
+
},
|
| 468 |
+
"50310": {
|
| 469 |
+
"content": "[unused25]",
|
| 470 |
+
"lstrip": false,
|
| 471 |
+
"normalized": true,
|
| 472 |
+
"rstrip": false,
|
| 473 |
+
"single_word": false,
|
| 474 |
+
"special": false
|
| 475 |
+
},
|
| 476 |
+
"50311": {
|
| 477 |
+
"content": "[unused26]",
|
| 478 |
+
"lstrip": false,
|
| 479 |
+
"normalized": true,
|
| 480 |
+
"rstrip": false,
|
| 481 |
+
"single_word": false,
|
| 482 |
+
"special": false
|
| 483 |
+
},
|
| 484 |
+
"50312": {
|
| 485 |
+
"content": "[unused27]",
|
| 486 |
+
"lstrip": false,
|
| 487 |
+
"normalized": true,
|
| 488 |
+
"rstrip": false,
|
| 489 |
+
"single_word": false,
|
| 490 |
+
"special": false
|
| 491 |
+
},
|
| 492 |
+
"50313": {
|
| 493 |
+
"content": "[unused28]",
|
| 494 |
+
"lstrip": false,
|
| 495 |
+
"normalized": true,
|
| 496 |
+
"rstrip": false,
|
| 497 |
+
"single_word": false,
|
| 498 |
+
"special": false
|
| 499 |
+
},
|
| 500 |
+
"50314": {
|
| 501 |
+
"content": "[unused29]",
|
| 502 |
+
"lstrip": false,
|
| 503 |
+
"normalized": true,
|
| 504 |
+
"rstrip": false,
|
| 505 |
+
"single_word": false,
|
| 506 |
+
"special": false
|
| 507 |
+
},
|
| 508 |
+
"50315": {
|
| 509 |
+
"content": "[unused30]",
|
| 510 |
+
"lstrip": false,
|
| 511 |
+
"normalized": true,
|
| 512 |
+
"rstrip": false,
|
| 513 |
+
"single_word": false,
|
| 514 |
+
"special": false
|
| 515 |
+
},
|
| 516 |
+
"50316": {
|
| 517 |
+
"content": "[unused31]",
|
| 518 |
+
"lstrip": false,
|
| 519 |
+
"normalized": true,
|
| 520 |
+
"rstrip": false,
|
| 521 |
+
"single_word": false,
|
| 522 |
+
"special": false
|
| 523 |
+
},
|
| 524 |
+
"50317": {
|
| 525 |
+
"content": "[unused32]",
|
| 526 |
+
"lstrip": false,
|
| 527 |
+
"normalized": true,
|
| 528 |
+
"rstrip": false,
|
| 529 |
+
"single_word": false,
|
| 530 |
+
"special": false
|
| 531 |
+
},
|
| 532 |
+
"50318": {
|
| 533 |
+
"content": "[unused33]",
|
| 534 |
+
"lstrip": false,
|
| 535 |
+
"normalized": true,
|
| 536 |
+
"rstrip": false,
|
| 537 |
+
"single_word": false,
|
| 538 |
+
"special": false
|
| 539 |
+
},
|
| 540 |
+
"50319": {
|
| 541 |
+
"content": "[unused34]",
|
| 542 |
+
"lstrip": false,
|
| 543 |
+
"normalized": true,
|
| 544 |
+
"rstrip": false,
|
| 545 |
+
"single_word": false,
|
| 546 |
+
"special": false
|
| 547 |
+
},
|
| 548 |
+
"50320": {
|
| 549 |
+
"content": "[unused35]",
|
| 550 |
+
"lstrip": false,
|
| 551 |
+
"normalized": true,
|
| 552 |
+
"rstrip": false,
|
| 553 |
+
"single_word": false,
|
| 554 |
+
"special": false
|
| 555 |
+
},
|
| 556 |
+
"50321": {
|
| 557 |
+
"content": "[unused36]",
|
| 558 |
+
"lstrip": false,
|
| 559 |
+
"normalized": true,
|
| 560 |
+
"rstrip": false,
|
| 561 |
+
"single_word": false,
|
| 562 |
+
"special": false
|
| 563 |
+
},
|
| 564 |
+
"50322": {
|
| 565 |
+
"content": "[unused37]",
|
| 566 |
+
"lstrip": false,
|
| 567 |
+
"normalized": true,
|
| 568 |
+
"rstrip": false,
|
| 569 |
+
"single_word": false,
|
| 570 |
+
"special": false
|
| 571 |
+
},
|
| 572 |
+
"50323": {
|
| 573 |
+
"content": "[unused38]",
|
| 574 |
+
"lstrip": false,
|
| 575 |
+
"normalized": true,
|
| 576 |
+
"rstrip": false,
|
| 577 |
+
"single_word": false,
|
| 578 |
+
"special": false
|
| 579 |
+
},
|
| 580 |
+
"50324": {
|
| 581 |
+
"content": "[unused39]",
|
| 582 |
+
"lstrip": false,
|
| 583 |
+
"normalized": true,
|
| 584 |
+
"rstrip": false,
|
| 585 |
+
"single_word": false,
|
| 586 |
+
"special": false
|
| 587 |
+
},
|
| 588 |
+
"50325": {
|
| 589 |
+
"content": "[unused40]",
|
| 590 |
+
"lstrip": false,
|
| 591 |
+
"normalized": true,
|
| 592 |
+
"rstrip": false,
|
| 593 |
+
"single_word": false,
|
| 594 |
+
"special": false
|
| 595 |
+
},
|
| 596 |
+
"50326": {
|
| 597 |
+
"content": "[unused41]",
|
| 598 |
+
"lstrip": false,
|
| 599 |
+
"normalized": true,
|
| 600 |
+
"rstrip": false,
|
| 601 |
+
"single_word": false,
|
| 602 |
+
"special": false
|
| 603 |
+
},
|
| 604 |
+
"50327": {
|
| 605 |
+
"content": "[unused42]",
|
| 606 |
+
"lstrip": false,
|
| 607 |
+
"normalized": true,
|
| 608 |
+
"rstrip": false,
|
| 609 |
+
"single_word": false,
|
| 610 |
+
"special": false
|
| 611 |
+
},
|
| 612 |
+
"50328": {
|
| 613 |
+
"content": "[unused43]",
|
| 614 |
+
"lstrip": false,
|
| 615 |
+
"normalized": true,
|
| 616 |
+
"rstrip": false,
|
| 617 |
+
"single_word": false,
|
| 618 |
+
"special": false
|
| 619 |
+
},
|
| 620 |
+
"50329": {
|
| 621 |
+
"content": "[unused44]",
|
| 622 |
+
"lstrip": false,
|
| 623 |
+
"normalized": true,
|
| 624 |
+
"rstrip": false,
|
| 625 |
+
"single_word": false,
|
| 626 |
+
"special": false
|
| 627 |
+
},
|
| 628 |
+
"50330": {
|
| 629 |
+
"content": "[unused45]",
|
| 630 |
+
"lstrip": false,
|
| 631 |
+
"normalized": true,
|
| 632 |
+
"rstrip": false,
|
| 633 |
+
"single_word": false,
|
| 634 |
+
"special": false
|
| 635 |
+
},
|
| 636 |
+
"50331": {
|
| 637 |
+
"content": "[unused46]",
|
| 638 |
+
"lstrip": false,
|
| 639 |
+
"normalized": true,
|
| 640 |
+
"rstrip": false,
|
| 641 |
+
"single_word": false,
|
| 642 |
+
"special": false
|
| 643 |
+
},
|
| 644 |
+
"50332": {
|
| 645 |
+
"content": "[unused47]",
|
| 646 |
+
"lstrip": false,
|
| 647 |
+
"normalized": true,
|
| 648 |
+
"rstrip": false,
|
| 649 |
+
"single_word": false,
|
| 650 |
+
"special": false
|
| 651 |
+
},
|
| 652 |
+
"50333": {
|
| 653 |
+
"content": "[unused48]",
|
| 654 |
+
"lstrip": false,
|
| 655 |
+
"normalized": true,
|
| 656 |
+
"rstrip": false,
|
| 657 |
+
"single_word": false,
|
| 658 |
+
"special": false
|
| 659 |
+
},
|
| 660 |
+
"50334": {
|
| 661 |
+
"content": "[unused49]",
|
| 662 |
+
"lstrip": false,
|
| 663 |
+
"normalized": true,
|
| 664 |
+
"rstrip": false,
|
| 665 |
+
"single_word": false,
|
| 666 |
+
"special": false
|
| 667 |
+
},
|
| 668 |
+
"50335": {
|
| 669 |
+
"content": "[unused50]",
|
| 670 |
+
"lstrip": false,
|
| 671 |
+
"normalized": true,
|
| 672 |
+
"rstrip": false,
|
| 673 |
+
"single_word": false,
|
| 674 |
+
"special": false
|
| 675 |
+
},
|
| 676 |
+
"50336": {
|
| 677 |
+
"content": "[unused51]",
|
| 678 |
+
"lstrip": false,
|
| 679 |
+
"normalized": true,
|
| 680 |
+
"rstrip": false,
|
| 681 |
+
"single_word": false,
|
| 682 |
+
"special": false
|
| 683 |
+
},
|
| 684 |
+
"50337": {
|
| 685 |
+
"content": "[unused52]",
|
| 686 |
+
"lstrip": false,
|
| 687 |
+
"normalized": true,
|
| 688 |
+
"rstrip": false,
|
| 689 |
+
"single_word": false,
|
| 690 |
+
"special": false
|
| 691 |
+
},
|
| 692 |
+
"50338": {
|
| 693 |
+
"content": "[unused53]",
|
| 694 |
+
"lstrip": false,
|
| 695 |
+
"normalized": true,
|
| 696 |
+
"rstrip": false,
|
| 697 |
+
"single_word": false,
|
| 698 |
+
"special": false
|
| 699 |
+
},
|
| 700 |
+
"50339": {
|
| 701 |
+
"content": "[unused54]",
|
| 702 |
+
"lstrip": false,
|
| 703 |
+
"normalized": true,
|
| 704 |
+
"rstrip": false,
|
| 705 |
+
"single_word": false,
|
| 706 |
+
"special": false
|
| 707 |
+
},
|
| 708 |
+
"50340": {
|
| 709 |
+
"content": "[unused55]",
|
| 710 |
+
"lstrip": false,
|
| 711 |
+
"normalized": true,
|
| 712 |
+
"rstrip": false,
|
| 713 |
+
"single_word": false,
|
| 714 |
+
"special": false
|
| 715 |
+
},
|
| 716 |
+
"50341": {
|
| 717 |
+
"content": "[unused56]",
|
| 718 |
+
"lstrip": false,
|
| 719 |
+
"normalized": true,
|
| 720 |
+
"rstrip": false,
|
| 721 |
+
"single_word": false,
|
| 722 |
+
"special": false
|
| 723 |
+
},
|
| 724 |
+
"50342": {
|
| 725 |
+
"content": "[unused57]",
|
| 726 |
+
"lstrip": false,
|
| 727 |
+
"normalized": true,
|
| 728 |
+
"rstrip": false,
|
| 729 |
+
"single_word": false,
|
| 730 |
+
"special": false
|
| 731 |
+
},
|
| 732 |
+
"50343": {
|
| 733 |
+
"content": "[unused58]",
|
| 734 |
+
"lstrip": false,
|
| 735 |
+
"normalized": true,
|
| 736 |
+
"rstrip": false,
|
| 737 |
+
"single_word": false,
|
| 738 |
+
"special": false
|
| 739 |
+
},
|
| 740 |
+
"50344": {
|
| 741 |
+
"content": "[unused59]",
|
| 742 |
+
"lstrip": false,
|
| 743 |
+
"normalized": true,
|
| 744 |
+
"rstrip": false,
|
| 745 |
+
"single_word": false,
|
| 746 |
+
"special": false
|
| 747 |
+
},
|
| 748 |
+
"50345": {
|
| 749 |
+
"content": "[unused60]",
|
| 750 |
+
"lstrip": false,
|
| 751 |
+
"normalized": true,
|
| 752 |
+
"rstrip": false,
|
| 753 |
+
"single_word": false,
|
| 754 |
+
"special": false
|
| 755 |
+
},
|
| 756 |
+
"50346": {
|
| 757 |
+
"content": "[unused61]",
|
| 758 |
+
"lstrip": false,
|
| 759 |
+
"normalized": true,
|
| 760 |
+
"rstrip": false,
|
| 761 |
+
"single_word": false,
|
| 762 |
+
"special": false
|
| 763 |
+
},
|
| 764 |
+
"50347": {
|
| 765 |
+
"content": "[unused62]",
|
| 766 |
+
"lstrip": false,
|
| 767 |
+
"normalized": true,
|
| 768 |
+
"rstrip": false,
|
| 769 |
+
"single_word": false,
|
| 770 |
+
"special": false
|
| 771 |
+
},
|
| 772 |
+
"50348": {
|
| 773 |
+
"content": "[unused63]",
|
| 774 |
+
"lstrip": false,
|
| 775 |
+
"normalized": true,
|
| 776 |
+
"rstrip": false,
|
| 777 |
+
"single_word": false,
|
| 778 |
+
"special": false
|
| 779 |
+
},
|
| 780 |
+
"50349": {
|
| 781 |
+
"content": "[unused64]",
|
| 782 |
+
"lstrip": false,
|
| 783 |
+
"normalized": true,
|
| 784 |
+
"rstrip": false,
|
| 785 |
+
"single_word": false,
|
| 786 |
+
"special": false
|
| 787 |
+
},
|
| 788 |
+
"50350": {
|
| 789 |
+
"content": "[unused65]",
|
| 790 |
+
"lstrip": false,
|
| 791 |
+
"normalized": true,
|
| 792 |
+
"rstrip": false,
|
| 793 |
+
"single_word": false,
|
| 794 |
+
"special": false
|
| 795 |
+
},
|
| 796 |
+
"50351": {
|
| 797 |
+
"content": "[unused66]",
|
| 798 |
+
"lstrip": false,
|
| 799 |
+
"normalized": true,
|
| 800 |
+
"rstrip": false,
|
| 801 |
+
"single_word": false,
|
| 802 |
+
"special": false
|
| 803 |
+
},
|
| 804 |
+
"50352": {
|
| 805 |
+
"content": "[unused67]",
|
| 806 |
+
"lstrip": false,
|
| 807 |
+
"normalized": true,
|
| 808 |
+
"rstrip": false,
|
| 809 |
+
"single_word": false,
|
| 810 |
+
"special": false
|
| 811 |
+
},
|
| 812 |
+
"50353": {
|
| 813 |
+
"content": "[unused68]",
|
| 814 |
+
"lstrip": false,
|
| 815 |
+
"normalized": true,
|
| 816 |
+
"rstrip": false,
|
| 817 |
+
"single_word": false,
|
| 818 |
+
"special": false
|
| 819 |
+
},
|
| 820 |
+
"50354": {
|
| 821 |
+
"content": "[unused69]",
|
| 822 |
+
"lstrip": false,
|
| 823 |
+
"normalized": true,
|
| 824 |
+
"rstrip": false,
|
| 825 |
+
"single_word": false,
|
| 826 |
+
"special": false
|
| 827 |
+
},
|
| 828 |
+
"50355": {
|
| 829 |
+
"content": "[unused70]",
|
| 830 |
+
"lstrip": false,
|
| 831 |
+
"normalized": true,
|
| 832 |
+
"rstrip": false,
|
| 833 |
+
"single_word": false,
|
| 834 |
+
"special": false
|
| 835 |
+
},
|
| 836 |
+
"50356": {
|
| 837 |
+
"content": "[unused71]",
|
| 838 |
+
"lstrip": false,
|
| 839 |
+
"normalized": true,
|
| 840 |
+
"rstrip": false,
|
| 841 |
+
"single_word": false,
|
| 842 |
+
"special": false
|
| 843 |
+
},
|
| 844 |
+
"50357": {
|
| 845 |
+
"content": "[unused72]",
|
| 846 |
+
"lstrip": false,
|
| 847 |
+
"normalized": true,
|
| 848 |
+
"rstrip": false,
|
| 849 |
+
"single_word": false,
|
| 850 |
+
"special": false
|
| 851 |
+
},
|
| 852 |
+
"50358": {
|
| 853 |
+
"content": "[unused73]",
|
| 854 |
+
"lstrip": false,
|
| 855 |
+
"normalized": true,
|
| 856 |
+
"rstrip": false,
|
| 857 |
+
"single_word": false,
|
| 858 |
+
"special": false
|
| 859 |
+
},
|
| 860 |
+
"50359": {
|
| 861 |
+
"content": "[unused74]",
|
| 862 |
+
"lstrip": false,
|
| 863 |
+
"normalized": true,
|
| 864 |
+
"rstrip": false,
|
| 865 |
+
"single_word": false,
|
| 866 |
+
"special": false
|
| 867 |
+
},
|
| 868 |
+
"50360": {
|
| 869 |
+
"content": "[unused75]",
|
| 870 |
+
"lstrip": false,
|
| 871 |
+
"normalized": true,
|
| 872 |
+
"rstrip": false,
|
| 873 |
+
"single_word": false,
|
| 874 |
+
"special": false
|
| 875 |
+
},
|
| 876 |
+
"50361": {
|
| 877 |
+
"content": "[unused76]",
|
| 878 |
+
"lstrip": false,
|
| 879 |
+
"normalized": true,
|
| 880 |
+
"rstrip": false,
|
| 881 |
+
"single_word": false,
|
| 882 |
+
"special": false
|
| 883 |
+
},
|
| 884 |
+
"50362": {
|
| 885 |
+
"content": "[unused77]",
|
| 886 |
+
"lstrip": false,
|
| 887 |
+
"normalized": true,
|
| 888 |
+
"rstrip": false,
|
| 889 |
+
"single_word": false,
|
| 890 |
+
"special": false
|
| 891 |
+
},
|
| 892 |
+
"50363": {
|
| 893 |
+
"content": "[unused78]",
|
| 894 |
+
"lstrip": false,
|
| 895 |
+
"normalized": true,
|
| 896 |
+
"rstrip": false,
|
| 897 |
+
"single_word": false,
|
| 898 |
+
"special": false
|
| 899 |
+
},
|
| 900 |
+
"50364": {
|
| 901 |
+
"content": "[unused79]",
|
| 902 |
+
"lstrip": false,
|
| 903 |
+
"normalized": true,
|
| 904 |
+
"rstrip": false,
|
| 905 |
+
"single_word": false,
|
| 906 |
+
"special": false
|
| 907 |
+
},
|
| 908 |
+
"50365": {
|
| 909 |
+
"content": "[unused80]",
|
| 910 |
+
"lstrip": false,
|
| 911 |
+
"normalized": true,
|
| 912 |
+
"rstrip": false,
|
| 913 |
+
"single_word": false,
|
| 914 |
+
"special": false
|
| 915 |
+
},
|
| 916 |
+
"50366": {
|
| 917 |
+
"content": "[unused81]",
|
| 918 |
+
"lstrip": false,
|
| 919 |
+
"normalized": true,
|
| 920 |
+
"rstrip": false,
|
| 921 |
+
"single_word": false,
|
| 922 |
+
"special": false
|
| 923 |
+
},
|
| 924 |
+
"50367": {
|
| 925 |
+
"content": "[unused82]",
|
| 926 |
+
"lstrip": false,
|
| 927 |
+
"normalized": true,
|
| 928 |
+
"rstrip": false,
|
| 929 |
+
"single_word": false,
|
| 930 |
+
"special": false
|
| 931 |
+
},
|
| 932 |
+
"50368": {
|
| 933 |
+
"content": "<s>",
|
| 934 |
+
"lstrip": false,
|
| 935 |
+
"normalized": false,
|
| 936 |
+
"rstrip": false,
|
| 937 |
+
"single_word": false,
|
| 938 |
+
"special": true
|
| 939 |
+
},
|
| 940 |
+
"50369": {
|
| 941 |
+
"content": "</s>",
|
| 942 |
+
"lstrip": false,
|
| 943 |
+
"normalized": false,
|
| 944 |
+
"rstrip": false,
|
| 945 |
+
"single_word": false,
|
| 946 |
+
"special": true
|
| 947 |
+
}
|
| 948 |
+
},
|
| 949 |
+
"bos_token": "<s>",
|
| 950 |
+
"clean_up_tokenization_spaces": true,
|
| 951 |
+
"cls_token": "[CLS]",
|
| 952 |
+
"eos_token": "</s>",
|
| 953 |
+
"errors": "replace",
|
| 954 |
+
"extra_special_tokens": {},
|
| 955 |
+
"mask_token": "[MASK]",
|
| 956 |
+
"model_input_names": [
|
| 957 |
+
"input_ids",
|
| 958 |
+
"attention_mask"
|
| 959 |
+
],
|
| 960 |
+
"model_max_length": 1024,
|
| 961 |
+
"pad_token": "[PAD]",
|
| 962 |
+
"sep_token": "[SEP]",
|
| 963 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 964 |
+
"trim_offsets": true,
|
| 965 |
+
"unk_token": "[UNK]"
|
| 966 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,986 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5487671335091459,
|
| 3 |
+
"best_model_checkpoint": "/mnt/data/bert24/fineweb_edu/checkpoints/5e-5_one_label/checkpoint-13149",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 13149,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.022815423226100844,
|
| 13 |
+
"grad_norm": 3.930772542953491,
|
| 14 |
+
"learning_rate": 1.1407711613050422e-06,
|
| 15 |
+
"loss": 0.8134,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04563084645220169,
|
| 20 |
+
"grad_norm": 3.383699655532837,
|
| 21 |
+
"learning_rate": 2.2815423226100845e-06,
|
| 22 |
+
"loss": 0.4827,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06844626967830253,
|
| 27 |
+
"grad_norm": 13.854891777038574,
|
| 28 |
+
"learning_rate": 3.4223134839151265e-06,
|
| 29 |
+
"loss": 0.386,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.09126169290440338,
|
| 34 |
+
"grad_norm": 4.105241775512695,
|
| 35 |
+
"learning_rate": 4.563084645220169e-06,
|
| 36 |
+
"loss": 0.3565,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.11407711613050422,
|
| 41 |
+
"grad_norm": 3.2210636138916016,
|
| 42 |
+
"learning_rate": 5.703855806525212e-06,
|
| 43 |
+
"loss": 0.3272,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.13689253935660506,
|
| 48 |
+
"grad_norm": 4.857998847961426,
|
| 49 |
+
"learning_rate": 6.844626967830253e-06,
|
| 50 |
+
"loss": 0.3297,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15970796258270592,
|
| 55 |
+
"grad_norm": 2.4987571239471436,
|
| 56 |
+
"learning_rate": 7.985398129135297e-06,
|
| 57 |
+
"loss": 0.3093,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.18252338580880675,
|
| 62 |
+
"grad_norm": 6.888661861419678,
|
| 63 |
+
"learning_rate": 9.126169290440338e-06,
|
| 64 |
+
"loss": 0.3108,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2053388090349076,
|
| 69 |
+
"grad_norm": 4.0510358810424805,
|
| 70 |
+
"learning_rate": 1.0266940451745379e-05,
|
| 71 |
+
"loss": 0.318,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.22815423226100845,
|
| 76 |
+
"grad_norm": 7.22244119644165,
|
| 77 |
+
"learning_rate": 1.1407711613050424e-05,
|
| 78 |
+
"loss": 0.293,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2509696554871093,
|
| 83 |
+
"grad_norm": 7.43681526184082,
|
| 84 |
+
"learning_rate": 1.2548482774355466e-05,
|
| 85 |
+
"loss": 0.2853,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2737850787132101,
|
| 90 |
+
"grad_norm": 3.81840443611145,
|
| 91 |
+
"learning_rate": 1.3689253935660506e-05,
|
| 92 |
+
"loss": 0.2784,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.296600501939311,
|
| 97 |
+
"grad_norm": 8.012810707092285,
|
| 98 |
+
"learning_rate": 1.4830025096965549e-05,
|
| 99 |
+
"loss": 0.2698,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.31941592516541184,
|
| 104 |
+
"grad_norm": 4.390589714050293,
|
| 105 |
+
"learning_rate": 1.5970796258270593e-05,
|
| 106 |
+
"loss": 0.2653,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.34223134839151265,
|
| 111 |
+
"grad_norm": 9.778414726257324,
|
| 112 |
+
"learning_rate": 1.7111567419575633e-05,
|
| 113 |
+
"loss": 0.2666,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3650467716176135,
|
| 118 |
+
"grad_norm": 3.7897677421569824,
|
| 119 |
+
"learning_rate": 1.8252338580880676e-05,
|
| 120 |
+
"loss": 0.259,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.38786219484371437,
|
| 125 |
+
"grad_norm": 3.3615636825561523,
|
| 126 |
+
"learning_rate": 1.939310974218572e-05,
|
| 127 |
+
"loss": 0.2522,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4106776180698152,
|
| 132 |
+
"grad_norm": 5.232314109802246,
|
| 133 |
+
"learning_rate": 2.0533880903490758e-05,
|
| 134 |
+
"loss": 0.2547,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.43349304129591604,
|
| 139 |
+
"grad_norm": 3.6110997200012207,
|
| 140 |
+
"learning_rate": 2.1674652064795804e-05,
|
| 141 |
+
"loss": 0.2594,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4563084645220169,
|
| 146 |
+
"grad_norm": 2.4445459842681885,
|
| 147 |
+
"learning_rate": 2.2815423226100847e-05,
|
| 148 |
+
"loss": 0.2537,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4791238877481177,
|
| 153 |
+
"grad_norm": 3.6951136589050293,
|
| 154 |
+
"learning_rate": 2.3956194387405887e-05,
|
| 155 |
+
"loss": 0.2518,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5019393109742186,
|
| 160 |
+
"grad_norm": 3.9381537437438965,
|
| 161 |
+
"learning_rate": 2.5096965548710933e-05,
|
| 162 |
+
"loss": 0.2516,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5247547342003194,
|
| 167 |
+
"grad_norm": 1.5301635265350342,
|
| 168 |
+
"learning_rate": 2.623773671001597e-05,
|
| 169 |
+
"loss": 0.2435,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5475701574264202,
|
| 174 |
+
"grad_norm": 4.029539585113525,
|
| 175 |
+
"learning_rate": 2.7378507871321012e-05,
|
| 176 |
+
"loss": 0.2467,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5703855806525211,
|
| 181 |
+
"grad_norm": 5.817471027374268,
|
| 182 |
+
"learning_rate": 2.8519279032626055e-05,
|
| 183 |
+
"loss": 0.2452,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.593201003878622,
|
| 188 |
+
"grad_norm": 3.2799909114837646,
|
| 189 |
+
"learning_rate": 2.9660050193931098e-05,
|
| 190 |
+
"loss": 0.2425,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6160164271047228,
|
| 195 |
+
"grad_norm": 1.8384263515472412,
|
| 196 |
+
"learning_rate": 3.080082135523614e-05,
|
| 197 |
+
"loss": 0.2371,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6388318503308237,
|
| 202 |
+
"grad_norm": 2.4290475845336914,
|
| 203 |
+
"learning_rate": 3.194159251654119e-05,
|
| 204 |
+
"loss": 0.2484,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6616472735569244,
|
| 209 |
+
"grad_norm": 1.5778276920318604,
|
| 210 |
+
"learning_rate": 3.3082363677846226e-05,
|
| 211 |
+
"loss": 0.2328,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6844626967830253,
|
| 216 |
+
"grad_norm": 2.872375249862671,
|
| 217 |
+
"learning_rate": 3.4223134839151266e-05,
|
| 218 |
+
"loss": 0.2546,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7072781200091262,
|
| 223 |
+
"grad_norm": 1.8962109088897705,
|
| 224 |
+
"learning_rate": 3.536390600045631e-05,
|
| 225 |
+
"loss": 0.236,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.730093543235227,
|
| 230 |
+
"grad_norm": 2.156942129135132,
|
| 231 |
+
"learning_rate": 3.650467716176135e-05,
|
| 232 |
+
"loss": 0.2366,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7529089664613279,
|
| 237 |
+
"grad_norm": 2.4454355239868164,
|
| 238 |
+
"learning_rate": 3.76454483230664e-05,
|
| 239 |
+
"loss": 0.2304,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7757243896874287,
|
| 244 |
+
"grad_norm": 4.029500961303711,
|
| 245 |
+
"learning_rate": 3.878621948437144e-05,
|
| 246 |
+
"loss": 0.2353,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7985398129135296,
|
| 251 |
+
"grad_norm": 1.7412891387939453,
|
| 252 |
+
"learning_rate": 3.9926990645676483e-05,
|
| 253 |
+
"loss": 0.2314,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8213552361396304,
|
| 258 |
+
"grad_norm": 2.626089572906494,
|
| 259 |
+
"learning_rate": 4.1067761806981516e-05,
|
| 260 |
+
"loss": 0.2327,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.8441706593657312,
|
| 265 |
+
"grad_norm": 1.2162595987319946,
|
| 266 |
+
"learning_rate": 4.220853296828656e-05,
|
| 267 |
+
"loss": 0.2367,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8669860825918321,
|
| 272 |
+
"grad_norm": 2.4519355297088623,
|
| 273 |
+
"learning_rate": 4.334930412959161e-05,
|
| 274 |
+
"loss": 0.2269,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8898015058179329,
|
| 279 |
+
"grad_norm": 1.2700366973876953,
|
| 280 |
+
"learning_rate": 4.449007529089665e-05,
|
| 281 |
+
"loss": 0.2334,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.9126169290440338,
|
| 286 |
+
"grad_norm": 2.6412575244903564,
|
| 287 |
+
"learning_rate": 4.5630846452201694e-05,
|
| 288 |
+
"loss": 0.2315,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.9354323522701347,
|
| 293 |
+
"grad_norm": 1.5039851665496826,
|
| 294 |
+
"learning_rate": 4.6771617613506734e-05,
|
| 295 |
+
"loss": 0.2335,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.9582477754962354,
|
| 300 |
+
"grad_norm": 5.099658012390137,
|
| 301 |
+
"learning_rate": 4.791238877481177e-05,
|
| 302 |
+
"loss": 0.2341,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9810631987223363,
|
| 307 |
+
"grad_norm": 3.0738842487335205,
|
| 308 |
+
"learning_rate": 4.905315993611681e-05,
|
| 309 |
+
"loss": 0.2375,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.0,
|
| 314 |
+
"eval_accuracy": 0.7342489784566673,
|
| 315 |
+
"eval_f1_macro": 0.49721066997853086,
|
| 316 |
+
"eval_loss": 0.2188662439584732,
|
| 317 |
+
"eval_precision": 0.6009352545578245,
|
| 318 |
+
"eval_recall": 0.4565070202147285,
|
| 319 |
+
"eval_runtime": 19.3025,
|
| 320 |
+
"eval_samples_per_second": 2421.601,
|
| 321 |
+
"eval_steps_per_second": 3.16,
|
| 322 |
+
"step": 4383
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.0038786219484372,
|
| 326 |
+
"grad_norm": 0.8934878706932068,
|
| 327 |
+
"learning_rate": 4.997845210028646e-05,
|
| 328 |
+
"loss": 0.2254,
|
| 329 |
+
"step": 4400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.0266940451745379,
|
| 333 |
+
"grad_norm": 2.297649621963501,
|
| 334 |
+
"learning_rate": 4.985169974903035e-05,
|
| 335 |
+
"loss": 0.2097,
|
| 336 |
+
"step": 4500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.0495094684006387,
|
| 340 |
+
"grad_norm": 1.5094428062438965,
|
| 341 |
+
"learning_rate": 4.972494739777423e-05,
|
| 342 |
+
"loss": 0.208,
|
| 343 |
+
"step": 4600
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.0723248916267396,
|
| 347 |
+
"grad_norm": 1.9187654256820679,
|
| 348 |
+
"learning_rate": 4.9598195046518116e-05,
|
| 349 |
+
"loss": 0.2003,
|
| 350 |
+
"step": 4700
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.0951403148528405,
|
| 354 |
+
"grad_norm": 0.9724093079566956,
|
| 355 |
+
"learning_rate": 4.9471442695262e-05,
|
| 356 |
+
"loss": 0.1976,
|
| 357 |
+
"step": 4800
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.1179557380789413,
|
| 361 |
+
"grad_norm": 1.6662517786026,
|
| 362 |
+
"learning_rate": 4.9344690344005885e-05,
|
| 363 |
+
"loss": 0.1982,
|
| 364 |
+
"step": 4900
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.1407711613050422,
|
| 368 |
+
"grad_norm": 0.7583943009376526,
|
| 369 |
+
"learning_rate": 4.921793799274977e-05,
|
| 370 |
+
"loss": 0.2014,
|
| 371 |
+
"step": 5000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.163586584531143,
|
| 375 |
+
"grad_norm": 3.5475401878356934,
|
| 376 |
+
"learning_rate": 4.9091185641493654e-05,
|
| 377 |
+
"loss": 0.1982,
|
| 378 |
+
"step": 5100
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.186402007757244,
|
| 382 |
+
"grad_norm": 1.741073727607727,
|
| 383 |
+
"learning_rate": 4.896443329023754e-05,
|
| 384 |
+
"loss": 0.2044,
|
| 385 |
+
"step": 5200
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.2092174309833448,
|
| 389 |
+
"grad_norm": 1.7129486799240112,
|
| 390 |
+
"learning_rate": 4.883768093898142e-05,
|
| 391 |
+
"loss": 0.2039,
|
| 392 |
+
"step": 5300
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.2320328542094456,
|
| 396 |
+
"grad_norm": 1.3080261945724487,
|
| 397 |
+
"learning_rate": 4.87109285877253e-05,
|
| 398 |
+
"loss": 0.1997,
|
| 399 |
+
"step": 5400
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.2548482774355465,
|
| 403 |
+
"grad_norm": 2.1321465969085693,
|
| 404 |
+
"learning_rate": 4.8584176236469186e-05,
|
| 405 |
+
"loss": 0.2013,
|
| 406 |
+
"step": 5500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.2776637006616474,
|
| 410 |
+
"grad_norm": 1.2857780456542969,
|
| 411 |
+
"learning_rate": 4.845742388521307e-05,
|
| 412 |
+
"loss": 0.2038,
|
| 413 |
+
"step": 5600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.3004791238877482,
|
| 417 |
+
"grad_norm": 2.0227644443511963,
|
| 418 |
+
"learning_rate": 4.8330671533956955e-05,
|
| 419 |
+
"loss": 0.1952,
|
| 420 |
+
"step": 5700
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.323294547113849,
|
| 424 |
+
"grad_norm": 1.627146601676941,
|
| 425 |
+
"learning_rate": 4.820391918270084e-05,
|
| 426 |
+
"loss": 0.1971,
|
| 427 |
+
"step": 5800
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.3461099703399497,
|
| 431 |
+
"grad_norm": 1.5856671333312988,
|
| 432 |
+
"learning_rate": 4.8077166831444724e-05,
|
| 433 |
+
"loss": 0.1966,
|
| 434 |
+
"step": 5900
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.3689253935660506,
|
| 438 |
+
"grad_norm": 0.9896683692932129,
|
| 439 |
+
"learning_rate": 4.795041448018861e-05,
|
| 440 |
+
"loss": 0.1949,
|
| 441 |
+
"step": 6000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.3917408167921514,
|
| 445 |
+
"grad_norm": 1.0868607759475708,
|
| 446 |
+
"learning_rate": 4.782366212893249e-05,
|
| 447 |
+
"loss": 0.1936,
|
| 448 |
+
"step": 6100
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.4145562400182523,
|
| 452 |
+
"grad_norm": 1.2213613986968994,
|
| 453 |
+
"learning_rate": 4.769690977767638e-05,
|
| 454 |
+
"loss": 0.1951,
|
| 455 |
+
"step": 6200
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.4373716632443532,
|
| 459 |
+
"grad_norm": 2.0876383781433105,
|
| 460 |
+
"learning_rate": 4.757015742642026e-05,
|
| 461 |
+
"loss": 0.1956,
|
| 462 |
+
"step": 6300
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.460187086470454,
|
| 466 |
+
"grad_norm": 3.948638916015625,
|
| 467 |
+
"learning_rate": 4.7443405075164146e-05,
|
| 468 |
+
"loss": 0.1999,
|
| 469 |
+
"step": 6400
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.483002509696555,
|
| 473 |
+
"grad_norm": 2.2048580646514893,
|
| 474 |
+
"learning_rate": 4.731665272390803e-05,
|
| 475 |
+
"loss": 0.1984,
|
| 476 |
+
"step": 6500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.5058179329226558,
|
| 480 |
+
"grad_norm": 1.1835227012634277,
|
| 481 |
+
"learning_rate": 4.7189900372651915e-05,
|
| 482 |
+
"loss": 0.1936,
|
| 483 |
+
"step": 6600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.5286333561487564,
|
| 487 |
+
"grad_norm": 1.2532591819763184,
|
| 488 |
+
"learning_rate": 4.70631480213958e-05,
|
| 489 |
+
"loss": 0.2015,
|
| 490 |
+
"step": 6700
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.5514487793748573,
|
| 494 |
+
"grad_norm": 1.1021713018417358,
|
| 495 |
+
"learning_rate": 4.6936395670139684e-05,
|
| 496 |
+
"loss": 0.1946,
|
| 497 |
+
"step": 6800
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.5742642026009581,
|
| 501 |
+
"grad_norm": 0.7517445087432861,
|
| 502 |
+
"learning_rate": 4.680964331888357e-05,
|
| 503 |
+
"loss": 0.2024,
|
| 504 |
+
"step": 6900
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.597079625827059,
|
| 508 |
+
"grad_norm": 1.0925252437591553,
|
| 509 |
+
"learning_rate": 4.668289096762745e-05,
|
| 510 |
+
"loss": 0.2004,
|
| 511 |
+
"step": 7000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.6198950490531598,
|
| 515 |
+
"grad_norm": 1.2946810722351074,
|
| 516 |
+
"learning_rate": 4.655613861637134e-05,
|
| 517 |
+
"loss": 0.1941,
|
| 518 |
+
"step": 7100
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.6427104722792607,
|
| 522 |
+
"grad_norm": 1.1611590385437012,
|
| 523 |
+
"learning_rate": 4.642938626511522e-05,
|
| 524 |
+
"loss": 0.1987,
|
| 525 |
+
"step": 7200
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.6655258955053616,
|
| 529 |
+
"grad_norm": 0.7706720232963562,
|
| 530 |
+
"learning_rate": 4.6302633913859107e-05,
|
| 531 |
+
"loss": 0.1977,
|
| 532 |
+
"step": 7300
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.6883413187314624,
|
| 536 |
+
"grad_norm": 2.32635760307312,
|
| 537 |
+
"learning_rate": 4.617588156260299e-05,
|
| 538 |
+
"loss": 0.2002,
|
| 539 |
+
"step": 7400
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.7111567419575633,
|
| 543 |
+
"grad_norm": 2.003863573074341,
|
| 544 |
+
"learning_rate": 4.6049129211346876e-05,
|
| 545 |
+
"loss": 0.1981,
|
| 546 |
+
"step": 7500
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.7339721651836641,
|
| 550 |
+
"grad_norm": 0.9495589137077332,
|
| 551 |
+
"learning_rate": 4.592237686009075e-05,
|
| 552 |
+
"loss": 0.1969,
|
| 553 |
+
"step": 7600
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.756787588409765,
|
| 557 |
+
"grad_norm": 1.1177550554275513,
|
| 558 |
+
"learning_rate": 4.579562450883464e-05,
|
| 559 |
+
"loss": 0.196,
|
| 560 |
+
"step": 7700
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.7796030116358659,
|
| 564 |
+
"grad_norm": 2.062082290649414,
|
| 565 |
+
"learning_rate": 4.566887215757852e-05,
|
| 566 |
+
"loss": 0.1979,
|
| 567 |
+
"step": 7800
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.8024184348619667,
|
| 571 |
+
"grad_norm": 1.265002965927124,
|
| 572 |
+
"learning_rate": 4.554211980632241e-05,
|
| 573 |
+
"loss": 0.1911,
|
| 574 |
+
"step": 7900
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 1.8252338580880676,
|
| 578 |
+
"grad_norm": 1.4993034601211548,
|
| 579 |
+
"learning_rate": 4.541536745506629e-05,
|
| 580 |
+
"loss": 0.1917,
|
| 581 |
+
"step": 8000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.8480492813141685,
|
| 585 |
+
"grad_norm": 0.7223235964775085,
|
| 586 |
+
"learning_rate": 4.5288615103810176e-05,
|
| 587 |
+
"loss": 0.1887,
|
| 588 |
+
"step": 8100
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.8708647045402693,
|
| 592 |
+
"grad_norm": 2.019066095352173,
|
| 593 |
+
"learning_rate": 4.516186275255406e-05,
|
| 594 |
+
"loss": 0.1963,
|
| 595 |
+
"step": 8200
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.8936801277663702,
|
| 599 |
+
"grad_norm": 1.8746905326843262,
|
| 600 |
+
"learning_rate": 4.5035110401297945e-05,
|
| 601 |
+
"loss": 0.191,
|
| 602 |
+
"step": 8300
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.916495550992471,
|
| 606 |
+
"grad_norm": 1.852623701095581,
|
| 607 |
+
"learning_rate": 4.490835805004183e-05,
|
| 608 |
+
"loss": 0.1923,
|
| 609 |
+
"step": 8400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.939310974218572,
|
| 613 |
+
"grad_norm": 1.784754753112793,
|
| 614 |
+
"learning_rate": 4.4781605698785714e-05,
|
| 615 |
+
"loss": 0.1931,
|
| 616 |
+
"step": 8500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 1.9621263974446728,
|
| 620 |
+
"grad_norm": 1.1549110412597656,
|
| 621 |
+
"learning_rate": 4.46548533475296e-05,
|
| 622 |
+
"loss": 0.196,
|
| 623 |
+
"step": 8600
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.9849418206707734,
|
| 627 |
+
"grad_norm": 1.884442925453186,
|
| 628 |
+
"learning_rate": 4.452810099627348e-05,
|
| 629 |
+
"loss": 0.1864,
|
| 630 |
+
"step": 8700
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.0,
|
| 634 |
+
"eval_accuracy": 0.7550221423528657,
|
| 635 |
+
"eval_f1_macro": 0.46991506497740926,
|
| 636 |
+
"eval_loss": 0.2009369283914566,
|
| 637 |
+
"eval_precision": 0.6326650393349825,
|
| 638 |
+
"eval_recall": 0.44929900989141963,
|
| 639 |
+
"eval_runtime": 18.7453,
|
| 640 |
+
"eval_samples_per_second": 2493.578,
|
| 641 |
+
"eval_steps_per_second": 3.254,
|
| 642 |
+
"step": 8766
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.0077572438968745,
|
| 646 |
+
"grad_norm": 0.6890222430229187,
|
| 647 |
+
"learning_rate": 4.440134864501737e-05,
|
| 648 |
+
"loss": 0.1749,
|
| 649 |
+
"step": 8800
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.0305726671229753,
|
| 653 |
+
"grad_norm": 0.682909369468689,
|
| 654 |
+
"learning_rate": 4.427459629376125e-05,
|
| 655 |
+
"loss": 0.1359,
|
| 656 |
+
"step": 8900
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.0533880903490758,
|
| 660 |
+
"grad_norm": 0.7629422545433044,
|
| 661 |
+
"learning_rate": 4.4147843942505136e-05,
|
| 662 |
+
"loss": 0.1367,
|
| 663 |
+
"step": 9000
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.0762035135751766,
|
| 667 |
+
"grad_norm": 0.9653803110122681,
|
| 668 |
+
"learning_rate": 4.402109159124902e-05,
|
| 669 |
+
"loss": 0.1358,
|
| 670 |
+
"step": 9100
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.0990189368012775,
|
| 674 |
+
"grad_norm": 1.0762972831726074,
|
| 675 |
+
"learning_rate": 4.3894339239992905e-05,
|
| 676 |
+
"loss": 0.136,
|
| 677 |
+
"step": 9200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 2.1218343600273784,
|
| 681 |
+
"grad_norm": 0.9004037976264954,
|
| 682 |
+
"learning_rate": 4.376758688873679e-05,
|
| 683 |
+
"loss": 0.1287,
|
| 684 |
+
"step": 9300
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.144649783253479,
|
| 688 |
+
"grad_norm": 1.2514455318450928,
|
| 689 |
+
"learning_rate": 4.3640834537480674e-05,
|
| 690 |
+
"loss": 0.1296,
|
| 691 |
+
"step": 9400
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.16746520647958,
|
| 695 |
+
"grad_norm": 0.9992089867591858,
|
| 696 |
+
"learning_rate": 4.351408218622456e-05,
|
| 697 |
+
"loss": 0.1343,
|
| 698 |
+
"step": 9500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.190280629705681,
|
| 702 |
+
"grad_norm": 1.090753197669983,
|
| 703 |
+
"learning_rate": 4.338732983496844e-05,
|
| 704 |
+
"loss": 0.1314,
|
| 705 |
+
"step": 9600
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.213096052931782,
|
| 709 |
+
"grad_norm": 1.0993868112564087,
|
| 710 |
+
"learning_rate": 4.326057748371232e-05,
|
| 711 |
+
"loss": 0.1294,
|
| 712 |
+
"step": 9700
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.2359114761578827,
|
| 716 |
+
"grad_norm": 7.720792770385742,
|
| 717 |
+
"learning_rate": 4.3133825132456206e-05,
|
| 718 |
+
"loss": 0.1311,
|
| 719 |
+
"step": 9800
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 2.2587268993839835,
|
| 723 |
+
"grad_norm": 0.968745768070221,
|
| 724 |
+
"learning_rate": 4.300707278120009e-05,
|
| 725 |
+
"loss": 0.1339,
|
| 726 |
+
"step": 9900
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 2.2815423226100844,
|
| 730 |
+
"grad_norm": 1.4330673217773438,
|
| 731 |
+
"learning_rate": 4.2880320429943975e-05,
|
| 732 |
+
"loss": 0.1354,
|
| 733 |
+
"step": 10000
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 2.3043577458361852,
|
| 737 |
+
"grad_norm": 1.1297446489334106,
|
| 738 |
+
"learning_rate": 4.275356807868786e-05,
|
| 739 |
+
"loss": 0.1343,
|
| 740 |
+
"step": 10100
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 2.327173169062286,
|
| 744 |
+
"grad_norm": 1.1260323524475098,
|
| 745 |
+
"learning_rate": 4.2626815727431744e-05,
|
| 746 |
+
"loss": 0.1324,
|
| 747 |
+
"step": 10200
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 2.349988592288387,
|
| 751 |
+
"grad_norm": 1.0184403657913208,
|
| 752 |
+
"learning_rate": 4.250006337617563e-05,
|
| 753 |
+
"loss": 0.13,
|
| 754 |
+
"step": 10300
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 2.372804015514488,
|
| 758 |
+
"grad_norm": 3.17570161819458,
|
| 759 |
+
"learning_rate": 4.237331102491951e-05,
|
| 760 |
+
"loss": 0.1316,
|
| 761 |
+
"step": 10400
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.3956194387405887,
|
| 765 |
+
"grad_norm": 0.9891201257705688,
|
| 766 |
+
"learning_rate": 4.22465586736634e-05,
|
| 767 |
+
"loss": 0.133,
|
| 768 |
+
"step": 10500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.4184348619666896,
|
| 772 |
+
"grad_norm": 1.0855801105499268,
|
| 773 |
+
"learning_rate": 4.211980632240728e-05,
|
| 774 |
+
"loss": 0.1362,
|
| 775 |
+
"step": 10600
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 2.4412502851927904,
|
| 779 |
+
"grad_norm": 1.94706392288208,
|
| 780 |
+
"learning_rate": 4.199305397115117e-05,
|
| 781 |
+
"loss": 0.1275,
|
| 782 |
+
"step": 10700
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 2.4640657084188913,
|
| 786 |
+
"grad_norm": 1.3414404392242432,
|
| 787 |
+
"learning_rate": 4.186630161989505e-05,
|
| 788 |
+
"loss": 0.1332,
|
| 789 |
+
"step": 10800
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.486881131644992,
|
| 793 |
+
"grad_norm": 1.5227530002593994,
|
| 794 |
+
"learning_rate": 4.1739549268638935e-05,
|
| 795 |
+
"loss": 0.1313,
|
| 796 |
+
"step": 10900
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.509696554871093,
|
| 800 |
+
"grad_norm": 1.4501299858093262,
|
| 801 |
+
"learning_rate": 4.161279691738282e-05,
|
| 802 |
+
"loss": 0.1368,
|
| 803 |
+
"step": 11000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.532511978097194,
|
| 807 |
+
"grad_norm": 1.6044068336486816,
|
| 808 |
+
"learning_rate": 4.1486044566126704e-05,
|
| 809 |
+
"loss": 0.1341,
|
| 810 |
+
"step": 11100
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 2.5553274013232947,
|
| 814 |
+
"grad_norm": 1.1524112224578857,
|
| 815 |
+
"learning_rate": 4.135929221487059e-05,
|
| 816 |
+
"loss": 0.134,
|
| 817 |
+
"step": 11200
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.5781428245493956,
|
| 821 |
+
"grad_norm": 2.4395620822906494,
|
| 822 |
+
"learning_rate": 4.123253986361447e-05,
|
| 823 |
+
"loss": 0.135,
|
| 824 |
+
"step": 11300
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.6009582477754964,
|
| 828 |
+
"grad_norm": 1.3260228633880615,
|
| 829 |
+
"learning_rate": 4.110578751235836e-05,
|
| 830 |
+
"loss": 0.1322,
|
| 831 |
+
"step": 11400
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.6237736710015973,
|
| 835 |
+
"grad_norm": 1.1100713014602661,
|
| 836 |
+
"learning_rate": 4.097903516110224e-05,
|
| 837 |
+
"loss": 0.1328,
|
| 838 |
+
"step": 11500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.646589094227698,
|
| 842 |
+
"grad_norm": 0.9172728061676025,
|
| 843 |
+
"learning_rate": 4.0852282809846127e-05,
|
| 844 |
+
"loss": 0.1286,
|
| 845 |
+
"step": 11600
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.669404517453799,
|
| 849 |
+
"grad_norm": 1.2171121835708618,
|
| 850 |
+
"learning_rate": 4.072553045859001e-05,
|
| 851 |
+
"loss": 0.1337,
|
| 852 |
+
"step": 11700
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.6922199406798994,
|
| 856 |
+
"grad_norm": 1.0674031972885132,
|
| 857 |
+
"learning_rate": 4.0598778107333896e-05,
|
| 858 |
+
"loss": 0.1324,
|
| 859 |
+
"step": 11800
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.7150353639060003,
|
| 863 |
+
"grad_norm": 1.2550758123397827,
|
| 864 |
+
"learning_rate": 4.047202575607777e-05,
|
| 865 |
+
"loss": 0.1342,
|
| 866 |
+
"step": 11900
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.737850787132101,
|
| 870 |
+
"grad_norm": 0.9326576590538025,
|
| 871 |
+
"learning_rate": 4.034527340482166e-05,
|
| 872 |
+
"loss": 0.1313,
|
| 873 |
+
"step": 12000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 2.760666210358202,
|
| 877 |
+
"grad_norm": 0.8683038949966431,
|
| 878 |
+
"learning_rate": 4.021852105356554e-05,
|
| 879 |
+
"loss": 0.1352,
|
| 880 |
+
"step": 12100
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 2.783481633584303,
|
| 884 |
+
"grad_norm": 1.100682020187378,
|
| 885 |
+
"learning_rate": 4.009176870230943e-05,
|
| 886 |
+
"loss": 0.1332,
|
| 887 |
+
"step": 12200
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 2.8062970568104038,
|
| 891 |
+
"grad_norm": 1.1301831007003784,
|
| 892 |
+
"learning_rate": 3.996501635105331e-05,
|
| 893 |
+
"loss": 0.1308,
|
| 894 |
+
"step": 12300
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 2.8291124800365046,
|
| 898 |
+
"grad_norm": 1.2631456851959229,
|
| 899 |
+
"learning_rate": 3.9838263999797196e-05,
|
| 900 |
+
"loss": 0.1324,
|
| 901 |
+
"step": 12400
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 2.8519279032626055,
|
| 905 |
+
"grad_norm": 1.0373079776763916,
|
| 906 |
+
"learning_rate": 3.971151164854108e-05,
|
| 907 |
+
"loss": 0.1311,
|
| 908 |
+
"step": 12500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.8747433264887063,
|
| 912 |
+
"grad_norm": 1.5447543859481812,
|
| 913 |
+
"learning_rate": 3.9584759297284965e-05,
|
| 914 |
+
"loss": 0.133,
|
| 915 |
+
"step": 12600
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 2.897558749714807,
|
| 919 |
+
"grad_norm": 1.0701638460159302,
|
| 920 |
+
"learning_rate": 3.945800694602885e-05,
|
| 921 |
+
"loss": 0.1313,
|
| 922 |
+
"step": 12700
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 2.920374172940908,
|
| 926 |
+
"grad_norm": 1.3061041831970215,
|
| 927 |
+
"learning_rate": 3.933125459477274e-05,
|
| 928 |
+
"loss": 0.1355,
|
| 929 |
+
"step": 12800
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 2.943189596167009,
|
| 933 |
+
"grad_norm": 1.4337469339370728,
|
| 934 |
+
"learning_rate": 3.920450224351662e-05,
|
| 935 |
+
"loss": 0.134,
|
| 936 |
+
"step": 12900
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 2.96600501939311,
|
| 940 |
+
"grad_norm": 1.2338460683822632,
|
| 941 |
+
"learning_rate": 3.90777498922605e-05,
|
| 942 |
+
"loss": 0.1334,
|
| 943 |
+
"step": 13000
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 2.9888204426192106,
|
| 947 |
+
"grad_norm": 1.1828727722167969,
|
| 948 |
+
"learning_rate": 3.895099754100439e-05,
|
| 949 |
+
"loss": 0.1374,
|
| 950 |
+
"step": 13100
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 3.0,
|
| 954 |
+
"eval_accuracy": 0.7612476734484308,
|
| 955 |
+
"eval_f1_macro": 0.5487671335091459,
|
| 956 |
+
"eval_loss": 0.19929102063179016,
|
| 957 |
+
"eval_precision": 0.6182830803008168,
|
| 958 |
+
"eval_recall": 0.5145348157777688,
|
| 959 |
+
"eval_runtime": 18.7521,
|
| 960 |
+
"eval_samples_per_second": 2492.68,
|
| 961 |
+
"eval_steps_per_second": 3.253,
|
| 962 |
+
"step": 13149
|
| 963 |
+
}
|
| 964 |
+
],
|
| 965 |
+
"logging_steps": 100,
|
| 966 |
+
"max_steps": 43830,
|
| 967 |
+
"num_input_tokens_seen": 0,
|
| 968 |
+
"num_train_epochs": 10,
|
| 969 |
+
"save_steps": 1000,
|
| 970 |
+
"stateful_callbacks": {
|
| 971 |
+
"TrainerControl": {
|
| 972 |
+
"args": {
|
| 973 |
+
"should_epoch_stop": false,
|
| 974 |
+
"should_evaluate": false,
|
| 975 |
+
"should_log": false,
|
| 976 |
+
"should_save": true,
|
| 977 |
+
"should_training_stop": false
|
| 978 |
+
},
|
| 979 |
+
"attributes": {}
|
| 980 |
+
}
|
| 981 |
+
},
|
| 982 |
+
"total_flos": 8.58047685121278e+17,
|
| 983 |
+
"train_batch_size": 16,
|
| 984 |
+
"trial_name": null,
|
| 985 |
+
"trial_params": null
|
| 986 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cddcf8fe9b83a256099ffa1f5beb31eba08146c9e921ccd9e0fc3b33817aec84
|
| 3 |
+
size 5432
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|