Upload folder using huggingface_hub
Browse files- checkpoint-7000/config.json +28 -0
- checkpoint-7000/model.safetensors +3 -0
- checkpoint-7000/optimizer.pt +3 -0
- checkpoint-7000/rng_state.pth +3 -0
- checkpoint-7000/scaler.pt +3 -0
- checkpoint-7000/scheduler.pt +3 -0
- checkpoint-7000/sentencepiece.bpe.model +3 -0
- checkpoint-7000/special_tokens_map.json +15 -0
- checkpoint-7000/tokenizer_config.json +56 -0
- checkpoint-7000/trainer_state.json +701 -0
- checkpoint-7000/training_args.bin +3 -0
- checkpoint-8000/config.json +28 -0
- checkpoint-8000/model.safetensors +3 -0
- checkpoint-8000/optimizer.pt +3 -0
- checkpoint-8000/rng_state.pth +3 -0
- checkpoint-8000/scaler.pt +3 -0
- checkpoint-8000/scheduler.pt +3 -0
- checkpoint-8000/sentencepiece.bpe.model +3 -0
- checkpoint-8000/special_tokens_map.json +15 -0
- checkpoint-8000/tokenizer_config.json +56 -0
- checkpoint-8000/trainer_state.json +795 -0
- checkpoint-8000/training_args.bin +3 -0
- checkpoint-8500/config.json +28 -0
- checkpoint-8500/model.safetensors +3 -0
- checkpoint-8500/optimizer.pt +3 -0
- checkpoint-8500/rng_state.pth +3 -0
- checkpoint-8500/scaler.pt +3 -0
- checkpoint-8500/scheduler.pt +3 -0
- checkpoint-8500/sentencepiece.bpe.model +3 -0
- checkpoint-8500/special_tokens_map.json +15 -0
- checkpoint-8500/tokenizer_config.json +56 -0
- checkpoint-8500/trainer_state.json +842 -0
- checkpoint-8500/training_args.bin +3 -0
- config.json +28 -0
- model.safetensors +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +15 -0
- test.py +32 -0
- tokenizer_config.json +56 -0
- training_args.bin +3 -0
checkpoint-7000/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"XLMRobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "single_label_classification",
|
| 24 |
+
"transformers_version": "4.57.3",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
checkpoint-7000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11581fcf59ed6effc51225813116b13565a436317e8381e00444ac3ce13d695e
|
| 3 |
+
size 1112205008
|
checkpoint-7000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbb0fcda173d1ae2f94dda534c96ed9a99e55c74784223db39e40e717eb61036
|
| 3 |
+
size 2224532875
|
checkpoint-7000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6dae84c904d708c130984bf940365e51dce8b1de42aca9c5e004b79c2964a51
|
| 3 |
+
size 14645
|
checkpoint-7000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67a4ac7aeda2e5e7d6451b732beb252e7b8b9aabde03ae3d1c084770435aed1b
|
| 3 |
+
size 1383
|
checkpoint-7000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4eaf7c4b5bbb5cc75cb44eda3f89bce1142fde81a196f090dfe5a1d7e6253341
|
| 3 |
+
size 1465
|
checkpoint-7000/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-7000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-7000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "<mask>",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"pad_token": "<pad>",
|
| 52 |
+
"sep_token": "</s>",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
checkpoint-7000/trainer_state.json
ADDED
|
@@ -0,0 +1,701 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 7000,
|
| 3 |
+
"best_metric": 0.9318026741133476,
|
| 4 |
+
"best_model_checkpoint": "./results/checkpoint-7000",
|
| 5 |
+
"epoch": 0.18054265965129476,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 7000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0025791808521613536,
|
| 14 |
+
"grad_norm": 8.55178451538086,
|
| 15 |
+
"learning_rate": 3.96e-06,
|
| 16 |
+
"loss": 0.6948,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.005158361704322707,
|
| 21 |
+
"grad_norm": 6.15341329574585,
|
| 22 |
+
"learning_rate": 7.960000000000002e-06,
|
| 23 |
+
"loss": 0.6256,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00773754255648406,
|
| 28 |
+
"grad_norm": 7.876453876495361,
|
| 29 |
+
"learning_rate": 1.196e-05,
|
| 30 |
+
"loss": 0.4996,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.010316723408645414,
|
| 35 |
+
"grad_norm": 25.46845817565918,
|
| 36 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 37 |
+
"loss": 0.4267,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.012895904260806768,
|
| 42 |
+
"grad_norm": 12.137995719909668,
|
| 43 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 44 |
+
"loss": 0.3932,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.012895904260806768,
|
| 49 |
+
"eval_accuracy": 0.8497975290810141,
|
| 50 |
+
"eval_f1": 0.8492073290894184,
|
| 51 |
+
"eval_loss": 0.36694180965423584,
|
| 52 |
+
"eval_precision": 0.8566847895515343,
|
| 53 |
+
"eval_recall": 0.8497975290810141,
|
| 54 |
+
"eval_runtime": 81.3998,
|
| 55 |
+
"eval_samples_per_second": 952.607,
|
| 56 |
+
"eval_steps_per_second": 59.546,
|
| 57 |
+
"step": 500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.01547508511296812,
|
| 61 |
+
"grad_norm": 9.444395065307617,
|
| 62 |
+
"learning_rate": 1.99829039165573e-05,
|
| 63 |
+
"loss": 0.3438,
|
| 64 |
+
"step": 600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.018054265965129476,
|
| 68 |
+
"grad_norm": 12.698025703430176,
|
| 69 |
+
"learning_rate": 1.9965635145403055e-05,
|
| 70 |
+
"loss": 0.3897,
|
| 71 |
+
"step": 700
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.02063344681729083,
|
| 75 |
+
"grad_norm": 7.648699760437012,
|
| 76 |
+
"learning_rate": 1.994836637424881e-05,
|
| 77 |
+
"loss": 0.3197,
|
| 78 |
+
"step": 800
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.02321262766945218,
|
| 82 |
+
"grad_norm": 11.770590782165527,
|
| 83 |
+
"learning_rate": 1.9931097603094565e-05,
|
| 84 |
+
"loss": 0.2641,
|
| 85 |
+
"step": 900
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.025791808521613536,
|
| 89 |
+
"grad_norm": 1.2174805402755737,
|
| 90 |
+
"learning_rate": 1.991382883194032e-05,
|
| 91 |
+
"loss": 0.3093,
|
| 92 |
+
"step": 1000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.025791808521613536,
|
| 96 |
+
"eval_accuracy": 0.8805163653245983,
|
| 97 |
+
"eval_f1": 0.8798423409034712,
|
| 98 |
+
"eval_loss": 0.29810118675231934,
|
| 99 |
+
"eval_precision": 0.8908826043373803,
|
| 100 |
+
"eval_recall": 0.8805163653245983,
|
| 101 |
+
"eval_runtime": 82.3661,
|
| 102 |
+
"eval_samples_per_second": 941.431,
|
| 103 |
+
"eval_steps_per_second": 58.847,
|
| 104 |
+
"step": 1000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02837098937377489,
|
| 108 |
+
"grad_norm": 35.91494369506836,
|
| 109 |
+
"learning_rate": 1.9896560060786075e-05,
|
| 110 |
+
"loss": 0.2701,
|
| 111 |
+
"step": 1100
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.03095017022593624,
|
| 115 |
+
"grad_norm": 10.724688529968262,
|
| 116 |
+
"learning_rate": 1.9879291289631834e-05,
|
| 117 |
+
"loss": 0.2723,
|
| 118 |
+
"step": 1200
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.0335293510780976,
|
| 122 |
+
"grad_norm": 11.826756477355957,
|
| 123 |
+
"learning_rate": 1.9862022518477585e-05,
|
| 124 |
+
"loss": 0.2669,
|
| 125 |
+
"step": 1300
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.03610853193025895,
|
| 129 |
+
"grad_norm": 13.958097457885742,
|
| 130 |
+
"learning_rate": 1.984475374732334e-05,
|
| 131 |
+
"loss": 0.2563,
|
| 132 |
+
"step": 1400
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.0386877127824203,
|
| 136 |
+
"grad_norm": 32.344303131103516,
|
| 137 |
+
"learning_rate": 1.98274849761691e-05,
|
| 138 |
+
"loss": 0.2717,
|
| 139 |
+
"step": 1500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.0386877127824203,
|
| 143 |
+
"eval_accuracy": 0.9098424079853499,
|
| 144 |
+
"eval_f1": 0.9098439407901873,
|
| 145 |
+
"eval_loss": 0.2351153939962387,
|
| 146 |
+
"eval_precision": 0.9100045564108226,
|
| 147 |
+
"eval_recall": 0.9098424079853499,
|
| 148 |
+
"eval_runtime": 80.9696,
|
| 149 |
+
"eval_samples_per_second": 957.669,
|
| 150 |
+
"eval_steps_per_second": 59.862,
|
| 151 |
+
"step": 1500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.04126689363458166,
|
| 155 |
+
"grad_norm": 41.87895584106445,
|
| 156 |
+
"learning_rate": 1.9810216205014854e-05,
|
| 157 |
+
"loss": 0.2775,
|
| 158 |
+
"step": 1600
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.04384607448674301,
|
| 162 |
+
"grad_norm": 9.029143333435059,
|
| 163 |
+
"learning_rate": 1.9792947433860606e-05,
|
| 164 |
+
"loss": 0.2747,
|
| 165 |
+
"step": 1700
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.04642525533890436,
|
| 169 |
+
"grad_norm": 0.5453509092330933,
|
| 170 |
+
"learning_rate": 1.9775678662706365e-05,
|
| 171 |
+
"loss": 0.2333,
|
| 172 |
+
"step": 1800
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.04900443619106572,
|
| 176 |
+
"grad_norm": 14.877126693725586,
|
| 177 |
+
"learning_rate": 1.975840989155212e-05,
|
| 178 |
+
"loss": 0.2335,
|
| 179 |
+
"step": 1900
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.05158361704322707,
|
| 183 |
+
"grad_norm": 9.152324676513672,
|
| 184 |
+
"learning_rate": 1.974114112039787e-05,
|
| 185 |
+
"loss": 0.2501,
|
| 186 |
+
"step": 2000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.05158361704322707,
|
| 190 |
+
"eval_accuracy": 0.915400686079802,
|
| 191 |
+
"eval_f1": 0.9153786857019529,
|
| 192 |
+
"eval_loss": 0.24714471399784088,
|
| 193 |
+
"eval_precision": 0.9155981210913413,
|
| 194 |
+
"eval_recall": 0.915400686079802,
|
| 195 |
+
"eval_runtime": 80.7839,
|
| 196 |
+
"eval_samples_per_second": 959.869,
|
| 197 |
+
"eval_steps_per_second": 60.0,
|
| 198 |
+
"step": 2000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.05416279789538842,
|
| 202 |
+
"grad_norm": 3.9494690895080566,
|
| 203 |
+
"learning_rate": 1.972387234924363e-05,
|
| 204 |
+
"loss": 0.2432,
|
| 205 |
+
"step": 2100
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.05674197874754978,
|
| 209 |
+
"grad_norm": 7.518190383911133,
|
| 210 |
+
"learning_rate": 1.9706603578089385e-05,
|
| 211 |
+
"loss": 0.2179,
|
| 212 |
+
"step": 2200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.05932115959971113,
|
| 216 |
+
"grad_norm": 55.069461822509766,
|
| 217 |
+
"learning_rate": 1.968933480693514e-05,
|
| 218 |
+
"loss": 0.2603,
|
| 219 |
+
"step": 2300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06190034045187248,
|
| 223 |
+
"grad_norm": 21.235607147216797,
|
| 224 |
+
"learning_rate": 1.9672066035780896e-05,
|
| 225 |
+
"loss": 0.2363,
|
| 226 |
+
"step": 2400
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.06447952130403384,
|
| 230 |
+
"grad_norm": 30.956710815429688,
|
| 231 |
+
"learning_rate": 1.965479726462665e-05,
|
| 232 |
+
"loss": 0.264,
|
| 233 |
+
"step": 2500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06447952130403384,
|
| 237 |
+
"eval_accuracy": 0.9174769802171726,
|
| 238 |
+
"eval_f1": 0.9174732815674826,
|
| 239 |
+
"eval_loss": 0.22409705817699432,
|
| 240 |
+
"eval_precision": 0.9178450257324412,
|
| 241 |
+
"eval_recall": 0.9174769802171726,
|
| 242 |
+
"eval_runtime": 82.4939,
|
| 243 |
+
"eval_samples_per_second": 939.972,
|
| 244 |
+
"eval_steps_per_second": 58.756,
|
| 245 |
+
"step": 2500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.0670587021561952,
|
| 249 |
+
"grad_norm": 3.0290513038635254,
|
| 250 |
+
"learning_rate": 1.9637528493472406e-05,
|
| 251 |
+
"loss": 0.2388,
|
| 252 |
+
"step": 2600
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.06963788300835655,
|
| 256 |
+
"grad_norm": 4.662931442260742,
|
| 257 |
+
"learning_rate": 1.962025972231816e-05,
|
| 258 |
+
"loss": 0.2132,
|
| 259 |
+
"step": 2700
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.0722170638605179,
|
| 263 |
+
"grad_norm": 31.94866180419922,
|
| 264 |
+
"learning_rate": 1.9602990951163916e-05,
|
| 265 |
+
"loss": 0.2576,
|
| 266 |
+
"step": 2800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.07479624471267925,
|
| 270 |
+
"grad_norm": 3.5957560539245605,
|
| 271 |
+
"learning_rate": 1.958572218000967e-05,
|
| 272 |
+
"loss": 0.2404,
|
| 273 |
+
"step": 2900
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.0773754255648406,
|
| 277 |
+
"grad_norm": 6.48631477355957,
|
| 278 |
+
"learning_rate": 1.9568453408855427e-05,
|
| 279 |
+
"loss": 0.2514,
|
| 280 |
+
"step": 3000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.0773754255648406,
|
| 284 |
+
"eval_accuracy": 0.9169998194526837,
|
| 285 |
+
"eval_f1": 0.9170017211060641,
|
| 286 |
+
"eval_loss": 0.21086551249027252,
|
| 287 |
+
"eval_precision": 0.9171348873594787,
|
| 288 |
+
"eval_recall": 0.9169998194526837,
|
| 289 |
+
"eval_runtime": 80.1992,
|
| 290 |
+
"eval_samples_per_second": 966.867,
|
| 291 |
+
"eval_steps_per_second": 60.437,
|
| 292 |
+
"step": 3000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.07995460641700196,
|
| 296 |
+
"grad_norm": 5.789911270141602,
|
| 297 |
+
"learning_rate": 1.9551184637701185e-05,
|
| 298 |
+
"loss": 0.2138,
|
| 299 |
+
"step": 3100
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.08253378726916331,
|
| 303 |
+
"grad_norm": 8.613458633422852,
|
| 304 |
+
"learning_rate": 1.9533915866546937e-05,
|
| 305 |
+
"loss": 0.2323,
|
| 306 |
+
"step": 3200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.08511296812132467,
|
| 310 |
+
"grad_norm": 5.5360822677612305,
|
| 311 |
+
"learning_rate": 1.9516647095392692e-05,
|
| 312 |
+
"loss": 0.2474,
|
| 313 |
+
"step": 3300
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08769214897348603,
|
| 317 |
+
"grad_norm": 15.3199462890625,
|
| 318 |
+
"learning_rate": 1.949937832423845e-05,
|
| 319 |
+
"loss": 0.1964,
|
| 320 |
+
"step": 3400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.09027132982564738,
|
| 324 |
+
"grad_norm": 0.23756052553653717,
|
| 325 |
+
"learning_rate": 1.9482109553084202e-05,
|
| 326 |
+
"loss": 0.2048,
|
| 327 |
+
"step": 3500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.09027132982564738,
|
| 331 |
+
"eval_accuracy": 0.9212555776224498,
|
| 332 |
+
"eval_f1": 0.9212388888583593,
|
| 333 |
+
"eval_loss": 0.217011496424675,
|
| 334 |
+
"eval_precision": 0.9220491698721403,
|
| 335 |
+
"eval_recall": 0.9212555776224498,
|
| 336 |
+
"eval_runtime": 82.2451,
|
| 337 |
+
"eval_samples_per_second": 942.816,
|
| 338 |
+
"eval_steps_per_second": 58.934,
|
| 339 |
+
"step": 3500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.09285051067780872,
|
| 343 |
+
"grad_norm": 3.4589858055114746,
|
| 344 |
+
"learning_rate": 1.9464840781929957e-05,
|
| 345 |
+
"loss": 0.2335,
|
| 346 |
+
"step": 3600
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.09542969152997008,
|
| 350 |
+
"grad_norm": 6.5932111740112305,
|
| 351 |
+
"learning_rate": 1.9447572010775716e-05,
|
| 352 |
+
"loss": 0.2239,
|
| 353 |
+
"step": 3700
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.09800887238213143,
|
| 357 |
+
"grad_norm": 2.6831107139587402,
|
| 358 |
+
"learning_rate": 1.943030323962147e-05,
|
| 359 |
+
"loss": 0.2049,
|
| 360 |
+
"step": 3800
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.10058805323429279,
|
| 364 |
+
"grad_norm": 26.725677490234375,
|
| 365 |
+
"learning_rate": 1.9413034468467223e-05,
|
| 366 |
+
"loss": 0.2404,
|
| 367 |
+
"step": 3900
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.10316723408645415,
|
| 371 |
+
"grad_norm": 26.966718673706055,
|
| 372 |
+
"learning_rate": 1.939576569731298e-05,
|
| 373 |
+
"loss": 0.241,
|
| 374 |
+
"step": 4000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.10316723408645415,
|
| 378 |
+
"eval_accuracy": 0.9104227386448634,
|
| 379 |
+
"eval_f1": 0.9102303033501772,
|
| 380 |
+
"eval_loss": 0.22579680383205414,
|
| 381 |
+
"eval_precision": 0.9150459252068764,
|
| 382 |
+
"eval_recall": 0.9104227386448634,
|
| 383 |
+
"eval_runtime": 81.0753,
|
| 384 |
+
"eval_samples_per_second": 956.419,
|
| 385 |
+
"eval_steps_per_second": 59.784,
|
| 386 |
+
"step": 4000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.1057464149386155,
|
| 390 |
+
"grad_norm": 17.290939331054688,
|
| 391 |
+
"learning_rate": 1.9378496926158737e-05,
|
| 392 |
+
"loss": 0.2298,
|
| 393 |
+
"step": 4100
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.10832559579077684,
|
| 397 |
+
"grad_norm": 8.844673156738281,
|
| 398 |
+
"learning_rate": 1.9361228155004492e-05,
|
| 399 |
+
"loss": 0.224,
|
| 400 |
+
"step": 4200
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.1109047766429382,
|
| 404 |
+
"grad_norm": 2.0837674140930176,
|
| 405 |
+
"learning_rate": 1.9343959383850247e-05,
|
| 406 |
+
"loss": 0.2367,
|
| 407 |
+
"step": 4300
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.11348395749509955,
|
| 411 |
+
"grad_norm": 7.639578819274902,
|
| 412 |
+
"learning_rate": 1.9326690612696002e-05,
|
| 413 |
+
"loss": 0.2194,
|
| 414 |
+
"step": 4400
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.11606313834726091,
|
| 418 |
+
"grad_norm": 9.39929485321045,
|
| 419 |
+
"learning_rate": 1.9309421841541757e-05,
|
| 420 |
+
"loss": 0.2297,
|
| 421 |
+
"step": 4500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.11606313834726091,
|
| 425 |
+
"eval_accuracy": 0.9185473678780531,
|
| 426 |
+
"eval_f1": 0.918488362019573,
|
| 427 |
+
"eval_loss": 0.24766957759857178,
|
| 428 |
+
"eval_precision": 0.9204437423697496,
|
| 429 |
+
"eval_recall": 0.9185473678780531,
|
| 430 |
+
"eval_runtime": 80.6546,
|
| 431 |
+
"eval_samples_per_second": 961.408,
|
| 432 |
+
"eval_steps_per_second": 60.096,
|
| 433 |
+
"step": 4500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.11864231919942227,
|
| 437 |
+
"grad_norm": 63.414676666259766,
|
| 438 |
+
"learning_rate": 1.9292153070387512e-05,
|
| 439 |
+
"loss": 0.2067,
|
| 440 |
+
"step": 4600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.12122150005158362,
|
| 444 |
+
"grad_norm": 27.895687103271484,
|
| 445 |
+
"learning_rate": 1.9274884299233268e-05,
|
| 446 |
+
"loss": 0.1975,
|
| 447 |
+
"step": 4700
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.12380068090374496,
|
| 451 |
+
"grad_norm": 5.536584377288818,
|
| 452 |
+
"learning_rate": 1.9257615528079023e-05,
|
| 453 |
+
"loss": 0.2141,
|
| 454 |
+
"step": 4800
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.12637986175590632,
|
| 458 |
+
"grad_norm": 18.6892147064209,
|
| 459 |
+
"learning_rate": 1.9240346756924778e-05,
|
| 460 |
+
"loss": 0.2549,
|
| 461 |
+
"step": 4900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.12895904260806768,
|
| 465 |
+
"grad_norm": 3.705787420272827,
|
| 466 |
+
"learning_rate": 1.9223077985770536e-05,
|
| 467 |
+
"loss": 0.2187,
|
| 468 |
+
"step": 5000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.12895904260806768,
|
| 472 |
+
"eval_accuracy": 0.9244409481313353,
|
| 473 |
+
"eval_f1": 0.9243951823390982,
|
| 474 |
+
"eval_loss": 0.22254011034965515,
|
| 475 |
+
"eval_precision": 0.9250643288957023,
|
| 476 |
+
"eval_recall": 0.9244409481313353,
|
| 477 |
+
"eval_runtime": 84.6134,
|
| 478 |
+
"eval_samples_per_second": 916.427,
|
| 479 |
+
"eval_steps_per_second": 57.284,
|
| 480 |
+
"step": 5000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.13153822346022903,
|
| 484 |
+
"grad_norm": 10.738173484802246,
|
| 485 |
+
"learning_rate": 1.9205809214616288e-05,
|
| 486 |
+
"loss": 0.2143,
|
| 487 |
+
"step": 5100
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.1341174043123904,
|
| 491 |
+
"grad_norm": 7.39179801940918,
|
| 492 |
+
"learning_rate": 1.9188540443462043e-05,
|
| 493 |
+
"loss": 0.205,
|
| 494 |
+
"step": 5200
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.13669658516455174,
|
| 498 |
+
"grad_norm": 14.402508735656738,
|
| 499 |
+
"learning_rate": 1.9171271672307802e-05,
|
| 500 |
+
"loss": 0.2027,
|
| 501 |
+
"step": 5300
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.1392757660167131,
|
| 505 |
+
"grad_norm": 8.077102661132812,
|
| 506 |
+
"learning_rate": 1.9154002901153554e-05,
|
| 507 |
+
"loss": 0.1909,
|
| 508 |
+
"step": 5400
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.14185494686887445,
|
| 512 |
+
"grad_norm": 2.036444664001465,
|
| 513 |
+
"learning_rate": 1.913673412999931e-05,
|
| 514 |
+
"loss": 0.2285,
|
| 515 |
+
"step": 5500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.14185494686887445,
|
| 519 |
+
"eval_accuracy": 0.9220164555982564,
|
| 520 |
+
"eval_f1": 0.9220066809051832,
|
| 521 |
+
"eval_loss": 0.2105225920677185,
|
| 522 |
+
"eval_precision": 0.9220853946451525,
|
| 523 |
+
"eval_recall": 0.9220164555982564,
|
| 524 |
+
"eval_runtime": 85.408,
|
| 525 |
+
"eval_samples_per_second": 907.901,
|
| 526 |
+
"eval_steps_per_second": 56.751,
|
| 527 |
+
"step": 5500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1444341277210358,
|
| 531 |
+
"grad_norm": 3.172539472579956,
|
| 532 |
+
"learning_rate": 1.9119465358845067e-05,
|
| 533 |
+
"loss": 0.1839,
|
| 534 |
+
"step": 5600
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.14701330857319717,
|
| 538 |
+
"grad_norm": 4.246190547943115,
|
| 539 |
+
"learning_rate": 1.9102196587690823e-05,
|
| 540 |
+
"loss": 0.2248,
|
| 541 |
+
"step": 5700
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1495924894253585,
|
| 545 |
+
"grad_norm": 9.676543235778809,
|
| 546 |
+
"learning_rate": 1.9084927816536574e-05,
|
| 547 |
+
"loss": 0.2003,
|
| 548 |
+
"step": 5800
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.15217167027751985,
|
| 552 |
+
"grad_norm": 14.093317031860352,
|
| 553 |
+
"learning_rate": 1.9067659045382333e-05,
|
| 554 |
+
"loss": 0.2283,
|
| 555 |
+
"step": 5900
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.1547508511296812,
|
| 559 |
+
"grad_norm": 6.108678817749023,
|
| 560 |
+
"learning_rate": 1.9050390274228088e-05,
|
| 561 |
+
"loss": 0.2001,
|
| 562 |
+
"step": 6000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.1547508511296812,
|
| 566 |
+
"eval_accuracy": 0.9288514611436383,
|
| 567 |
+
"eval_f1": 0.9288285810713337,
|
| 568 |
+
"eval_loss": 0.19547687470912933,
|
| 569 |
+
"eval_precision": 0.9291294558172835,
|
| 570 |
+
"eval_recall": 0.9288514611436383,
|
| 571 |
+
"eval_runtime": 85.1832,
|
| 572 |
+
"eval_samples_per_second": 910.297,
|
| 573 |
+
"eval_steps_per_second": 56.901,
|
| 574 |
+
"step": 6000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.15733003198184256,
|
| 578 |
+
"grad_norm": 25.980791091918945,
|
| 579 |
+
"learning_rate": 1.9033121503073843e-05,
|
| 580 |
+
"loss": 0.2045,
|
| 581 |
+
"step": 6100
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.15990921283400392,
|
| 585 |
+
"grad_norm": 3.697967767715454,
|
| 586 |
+
"learning_rate": 1.90158527319196e-05,
|
| 587 |
+
"loss": 0.1921,
|
| 588 |
+
"step": 6200
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.16248839368616527,
|
| 592 |
+
"grad_norm": 6.734536647796631,
|
| 593 |
+
"learning_rate": 1.8998583960765353e-05,
|
| 594 |
+
"loss": 0.2398,
|
| 595 |
+
"step": 6300
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.16506757453832663,
|
| 599 |
+
"grad_norm": 4.171863555908203,
|
| 600 |
+
"learning_rate": 1.898131518961111e-05,
|
| 601 |
+
"loss": 0.1851,
|
| 602 |
+
"step": 6400
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.16764675539048798,
|
| 606 |
+
"grad_norm": 8.22290325164795,
|
| 607 |
+
"learning_rate": 1.8964046418456864e-05,
|
| 608 |
+
"loss": 0.2048,
|
| 609 |
+
"step": 6500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.16764675539048798,
|
| 613 |
+
"eval_accuracy": 0.9237316551030409,
|
| 614 |
+
"eval_f1": 0.9237108101584136,
|
| 615 |
+
"eval_loss": 0.21942158043384552,
|
| 616 |
+
"eval_precision": 0.9246712025171661,
|
| 617 |
+
"eval_recall": 0.9237316551030409,
|
| 618 |
+
"eval_runtime": 87.4195,
|
| 619 |
+
"eval_samples_per_second": 887.01,
|
| 620 |
+
"eval_steps_per_second": 55.445,
|
| 621 |
+
"step": 6500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.17022593624264934,
|
| 625 |
+
"grad_norm": 1.4064490795135498,
|
| 626 |
+
"learning_rate": 1.894677764730262e-05,
|
| 627 |
+
"loss": 0.2192,
|
| 628 |
+
"step": 6600
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.1728051170948107,
|
| 632 |
+
"grad_norm": 0.8423879146575928,
|
| 633 |
+
"learning_rate": 1.8929508876148374e-05,
|
| 634 |
+
"loss": 0.2296,
|
| 635 |
+
"step": 6700
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.17538429794697205,
|
| 639 |
+
"grad_norm": 2.1388964653015137,
|
| 640 |
+
"learning_rate": 1.891224010499413e-05,
|
| 641 |
+
"loss": 0.2397,
|
| 642 |
+
"step": 6800
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.1779634787991334,
|
| 646 |
+
"grad_norm": 6.909358501434326,
|
| 647 |
+
"learning_rate": 1.8894971333839888e-05,
|
| 648 |
+
"loss": 0.1902,
|
| 649 |
+
"step": 6900
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.18054265965129476,
|
| 653 |
+
"grad_norm": 1.044767141342163,
|
| 654 |
+
"learning_rate": 1.887770256268564e-05,
|
| 655 |
+
"loss": 0.1931,
|
| 656 |
+
"step": 7000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.18054265965129476,
|
| 660 |
+
"eval_accuracy": 0.9318046993887183,
|
| 661 |
+
"eval_f1": 0.9318026741133476,
|
| 662 |
+
"eval_loss": 0.18849702179431915,
|
| 663 |
+
"eval_precision": 0.9318098201446341,
|
| 664 |
+
"eval_recall": 0.9318046993887183,
|
| 665 |
+
"eval_runtime": 85.7797,
|
| 666 |
+
"eval_samples_per_second": 903.967,
|
| 667 |
+
"eval_steps_per_second": 56.505,
|
| 668 |
+
"step": 7000
|
| 669 |
+
}
|
| 670 |
+
],
|
| 671 |
+
"logging_steps": 100,
|
| 672 |
+
"max_steps": 116316,
|
| 673 |
+
"num_input_tokens_seen": 0,
|
| 674 |
+
"num_train_epochs": 3,
|
| 675 |
+
"save_steps": 500,
|
| 676 |
+
"stateful_callbacks": {
|
| 677 |
+
"EarlyStoppingCallback": {
|
| 678 |
+
"args": {
|
| 679 |
+
"early_stopping_patience": 3,
|
| 680 |
+
"early_stopping_threshold": 0.0
|
| 681 |
+
},
|
| 682 |
+
"attributes": {
|
| 683 |
+
"early_stopping_patience_counter": 0
|
| 684 |
+
}
|
| 685 |
+
},
|
| 686 |
+
"TrainerControl": {
|
| 687 |
+
"args": {
|
| 688 |
+
"should_epoch_stop": false,
|
| 689 |
+
"should_evaluate": false,
|
| 690 |
+
"should_log": false,
|
| 691 |
+
"should_save": true,
|
| 692 |
+
"should_training_stop": false
|
| 693 |
+
},
|
| 694 |
+
"attributes": {}
|
| 695 |
+
}
|
| 696 |
+
},
|
| 697 |
+
"total_flos": 2.534886729544608e+16,
|
| 698 |
+
"train_batch_size": 16,
|
| 699 |
+
"trial_name": null,
|
| 700 |
+
"trial_params": null
|
| 701 |
+
}
|
checkpoint-7000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce8abd265ced130b3a60da4969f0ccef6e504e9f0ee20e35110ba851d1b32cb3
|
| 3 |
+
size 5777
|
checkpoint-8000/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"XLMRobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "single_label_classification",
|
| 24 |
+
"transformers_version": "4.57.3",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
checkpoint-8000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bf7bf3f6d9df4b677428e8787a2e9c23a1eed357393a2749bd1b331d815c577
|
| 3 |
+
size 1112205008
|
checkpoint-8000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b16044d5b6739847bb763d6e7b4df2ceebc97beff514912aa55dd7766cb9491c
|
| 3 |
+
size 2224532875
|
checkpoint-8000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7111253787a9df18bde957fc589d600eff7f44793b5e3bf64a2e32b4057938f9
|
| 3 |
+
size 14645
|
checkpoint-8000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:922ff24c213c3bda10e2eb4a4d8d5cc817d8cf7cca674392f4fd327654134e8c
|
| 3 |
+
size 1383
|
checkpoint-8000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d028fa8da82e9494cd9148b8fe1607d2f58300c39ad900a04937564f7c4a6fc6
|
| 3 |
+
size 1465
|
checkpoint-8000/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-8000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-8000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "<mask>",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"pad_token": "<pad>",
|
| 52 |
+
"sep_token": "</s>",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
checkpoint-8000/trainer_state.json
ADDED
|
@@ -0,0 +1,795 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 7000,
|
| 3 |
+
"best_metric": 0.9318026741133476,
|
| 4 |
+
"best_model_checkpoint": "./results/checkpoint-7000",
|
| 5 |
+
"epoch": 0.2063344681729083,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 8000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0025791808521613536,
|
| 14 |
+
"grad_norm": 8.55178451538086,
|
| 15 |
+
"learning_rate": 3.96e-06,
|
| 16 |
+
"loss": 0.6948,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.005158361704322707,
|
| 21 |
+
"grad_norm": 6.15341329574585,
|
| 22 |
+
"learning_rate": 7.960000000000002e-06,
|
| 23 |
+
"loss": 0.6256,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00773754255648406,
|
| 28 |
+
"grad_norm": 7.876453876495361,
|
| 29 |
+
"learning_rate": 1.196e-05,
|
| 30 |
+
"loss": 0.4996,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.010316723408645414,
|
| 35 |
+
"grad_norm": 25.46845817565918,
|
| 36 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 37 |
+
"loss": 0.4267,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.012895904260806768,
|
| 42 |
+
"grad_norm": 12.137995719909668,
|
| 43 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 44 |
+
"loss": 0.3932,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.012895904260806768,
|
| 49 |
+
"eval_accuracy": 0.8497975290810141,
|
| 50 |
+
"eval_f1": 0.8492073290894184,
|
| 51 |
+
"eval_loss": 0.36694180965423584,
|
| 52 |
+
"eval_precision": 0.8566847895515343,
|
| 53 |
+
"eval_recall": 0.8497975290810141,
|
| 54 |
+
"eval_runtime": 81.3998,
|
| 55 |
+
"eval_samples_per_second": 952.607,
|
| 56 |
+
"eval_steps_per_second": 59.546,
|
| 57 |
+
"step": 500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.01547508511296812,
|
| 61 |
+
"grad_norm": 9.444395065307617,
|
| 62 |
+
"learning_rate": 1.99829039165573e-05,
|
| 63 |
+
"loss": 0.3438,
|
| 64 |
+
"step": 600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.018054265965129476,
|
| 68 |
+
"grad_norm": 12.698025703430176,
|
| 69 |
+
"learning_rate": 1.9965635145403055e-05,
|
| 70 |
+
"loss": 0.3897,
|
| 71 |
+
"step": 700
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.02063344681729083,
|
| 75 |
+
"grad_norm": 7.648699760437012,
|
| 76 |
+
"learning_rate": 1.994836637424881e-05,
|
| 77 |
+
"loss": 0.3197,
|
| 78 |
+
"step": 800
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.02321262766945218,
|
| 82 |
+
"grad_norm": 11.770590782165527,
|
| 83 |
+
"learning_rate": 1.9931097603094565e-05,
|
| 84 |
+
"loss": 0.2641,
|
| 85 |
+
"step": 900
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.025791808521613536,
|
| 89 |
+
"grad_norm": 1.2174805402755737,
|
| 90 |
+
"learning_rate": 1.991382883194032e-05,
|
| 91 |
+
"loss": 0.3093,
|
| 92 |
+
"step": 1000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.025791808521613536,
|
| 96 |
+
"eval_accuracy": 0.8805163653245983,
|
| 97 |
+
"eval_f1": 0.8798423409034712,
|
| 98 |
+
"eval_loss": 0.29810118675231934,
|
| 99 |
+
"eval_precision": 0.8908826043373803,
|
| 100 |
+
"eval_recall": 0.8805163653245983,
|
| 101 |
+
"eval_runtime": 82.3661,
|
| 102 |
+
"eval_samples_per_second": 941.431,
|
| 103 |
+
"eval_steps_per_second": 58.847,
|
| 104 |
+
"step": 1000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02837098937377489,
|
| 108 |
+
"grad_norm": 35.91494369506836,
|
| 109 |
+
"learning_rate": 1.9896560060786075e-05,
|
| 110 |
+
"loss": 0.2701,
|
| 111 |
+
"step": 1100
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.03095017022593624,
|
| 115 |
+
"grad_norm": 10.724688529968262,
|
| 116 |
+
"learning_rate": 1.9879291289631834e-05,
|
| 117 |
+
"loss": 0.2723,
|
| 118 |
+
"step": 1200
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.0335293510780976,
|
| 122 |
+
"grad_norm": 11.826756477355957,
|
| 123 |
+
"learning_rate": 1.9862022518477585e-05,
|
| 124 |
+
"loss": 0.2669,
|
| 125 |
+
"step": 1300
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.03610853193025895,
|
| 129 |
+
"grad_norm": 13.958097457885742,
|
| 130 |
+
"learning_rate": 1.984475374732334e-05,
|
| 131 |
+
"loss": 0.2563,
|
| 132 |
+
"step": 1400
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.0386877127824203,
|
| 136 |
+
"grad_norm": 32.344303131103516,
|
| 137 |
+
"learning_rate": 1.98274849761691e-05,
|
| 138 |
+
"loss": 0.2717,
|
| 139 |
+
"step": 1500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.0386877127824203,
|
| 143 |
+
"eval_accuracy": 0.9098424079853499,
|
| 144 |
+
"eval_f1": 0.9098439407901873,
|
| 145 |
+
"eval_loss": 0.2351153939962387,
|
| 146 |
+
"eval_precision": 0.9100045564108226,
|
| 147 |
+
"eval_recall": 0.9098424079853499,
|
| 148 |
+
"eval_runtime": 80.9696,
|
| 149 |
+
"eval_samples_per_second": 957.669,
|
| 150 |
+
"eval_steps_per_second": 59.862,
|
| 151 |
+
"step": 1500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.04126689363458166,
|
| 155 |
+
"grad_norm": 41.87895584106445,
|
| 156 |
+
"learning_rate": 1.9810216205014854e-05,
|
| 157 |
+
"loss": 0.2775,
|
| 158 |
+
"step": 1600
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.04384607448674301,
|
| 162 |
+
"grad_norm": 9.029143333435059,
|
| 163 |
+
"learning_rate": 1.9792947433860606e-05,
|
| 164 |
+
"loss": 0.2747,
|
| 165 |
+
"step": 1700
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.04642525533890436,
|
| 169 |
+
"grad_norm": 0.5453509092330933,
|
| 170 |
+
"learning_rate": 1.9775678662706365e-05,
|
| 171 |
+
"loss": 0.2333,
|
| 172 |
+
"step": 1800
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.04900443619106572,
|
| 176 |
+
"grad_norm": 14.877126693725586,
|
| 177 |
+
"learning_rate": 1.975840989155212e-05,
|
| 178 |
+
"loss": 0.2335,
|
| 179 |
+
"step": 1900
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.05158361704322707,
|
| 183 |
+
"grad_norm": 9.152324676513672,
|
| 184 |
+
"learning_rate": 1.974114112039787e-05,
|
| 185 |
+
"loss": 0.2501,
|
| 186 |
+
"step": 2000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.05158361704322707,
|
| 190 |
+
"eval_accuracy": 0.915400686079802,
|
| 191 |
+
"eval_f1": 0.9153786857019529,
|
| 192 |
+
"eval_loss": 0.24714471399784088,
|
| 193 |
+
"eval_precision": 0.9155981210913413,
|
| 194 |
+
"eval_recall": 0.915400686079802,
|
| 195 |
+
"eval_runtime": 80.7839,
|
| 196 |
+
"eval_samples_per_second": 959.869,
|
| 197 |
+
"eval_steps_per_second": 60.0,
|
| 198 |
+
"step": 2000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.05416279789538842,
|
| 202 |
+
"grad_norm": 3.9494690895080566,
|
| 203 |
+
"learning_rate": 1.972387234924363e-05,
|
| 204 |
+
"loss": 0.2432,
|
| 205 |
+
"step": 2100
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.05674197874754978,
|
| 209 |
+
"grad_norm": 7.518190383911133,
|
| 210 |
+
"learning_rate": 1.9706603578089385e-05,
|
| 211 |
+
"loss": 0.2179,
|
| 212 |
+
"step": 2200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.05932115959971113,
|
| 216 |
+
"grad_norm": 55.069461822509766,
|
| 217 |
+
"learning_rate": 1.968933480693514e-05,
|
| 218 |
+
"loss": 0.2603,
|
| 219 |
+
"step": 2300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06190034045187248,
|
| 223 |
+
"grad_norm": 21.235607147216797,
|
| 224 |
+
"learning_rate": 1.9672066035780896e-05,
|
| 225 |
+
"loss": 0.2363,
|
| 226 |
+
"step": 2400
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.06447952130403384,
|
| 230 |
+
"grad_norm": 30.956710815429688,
|
| 231 |
+
"learning_rate": 1.965479726462665e-05,
|
| 232 |
+
"loss": 0.264,
|
| 233 |
+
"step": 2500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06447952130403384,
|
| 237 |
+
"eval_accuracy": 0.9174769802171726,
|
| 238 |
+
"eval_f1": 0.9174732815674826,
|
| 239 |
+
"eval_loss": 0.22409705817699432,
|
| 240 |
+
"eval_precision": 0.9178450257324412,
|
| 241 |
+
"eval_recall": 0.9174769802171726,
|
| 242 |
+
"eval_runtime": 82.4939,
|
| 243 |
+
"eval_samples_per_second": 939.972,
|
| 244 |
+
"eval_steps_per_second": 58.756,
|
| 245 |
+
"step": 2500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.0670587021561952,
|
| 249 |
+
"grad_norm": 3.0290513038635254,
|
| 250 |
+
"learning_rate": 1.9637528493472406e-05,
|
| 251 |
+
"loss": 0.2388,
|
| 252 |
+
"step": 2600
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.06963788300835655,
|
| 256 |
+
"grad_norm": 4.662931442260742,
|
| 257 |
+
"learning_rate": 1.962025972231816e-05,
|
| 258 |
+
"loss": 0.2132,
|
| 259 |
+
"step": 2700
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.0722170638605179,
|
| 263 |
+
"grad_norm": 31.94866180419922,
|
| 264 |
+
"learning_rate": 1.9602990951163916e-05,
|
| 265 |
+
"loss": 0.2576,
|
| 266 |
+
"step": 2800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.07479624471267925,
|
| 270 |
+
"grad_norm": 3.5957560539245605,
|
| 271 |
+
"learning_rate": 1.958572218000967e-05,
|
| 272 |
+
"loss": 0.2404,
|
| 273 |
+
"step": 2900
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.0773754255648406,
|
| 277 |
+
"grad_norm": 6.48631477355957,
|
| 278 |
+
"learning_rate": 1.9568453408855427e-05,
|
| 279 |
+
"loss": 0.2514,
|
| 280 |
+
"step": 3000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.0773754255648406,
|
| 284 |
+
"eval_accuracy": 0.9169998194526837,
|
| 285 |
+
"eval_f1": 0.9170017211060641,
|
| 286 |
+
"eval_loss": 0.21086551249027252,
|
| 287 |
+
"eval_precision": 0.9171348873594787,
|
| 288 |
+
"eval_recall": 0.9169998194526837,
|
| 289 |
+
"eval_runtime": 80.1992,
|
| 290 |
+
"eval_samples_per_second": 966.867,
|
| 291 |
+
"eval_steps_per_second": 60.437,
|
| 292 |
+
"step": 3000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.07995460641700196,
|
| 296 |
+
"grad_norm": 5.789911270141602,
|
| 297 |
+
"learning_rate": 1.9551184637701185e-05,
|
| 298 |
+
"loss": 0.2138,
|
| 299 |
+
"step": 3100
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.08253378726916331,
|
| 303 |
+
"grad_norm": 8.613458633422852,
|
| 304 |
+
"learning_rate": 1.9533915866546937e-05,
|
| 305 |
+
"loss": 0.2323,
|
| 306 |
+
"step": 3200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.08511296812132467,
|
| 310 |
+
"grad_norm": 5.5360822677612305,
|
| 311 |
+
"learning_rate": 1.9516647095392692e-05,
|
| 312 |
+
"loss": 0.2474,
|
| 313 |
+
"step": 3300
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08769214897348603,
|
| 317 |
+
"grad_norm": 15.3199462890625,
|
| 318 |
+
"learning_rate": 1.949937832423845e-05,
|
| 319 |
+
"loss": 0.1964,
|
| 320 |
+
"step": 3400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.09027132982564738,
|
| 324 |
+
"grad_norm": 0.23756052553653717,
|
| 325 |
+
"learning_rate": 1.9482109553084202e-05,
|
| 326 |
+
"loss": 0.2048,
|
| 327 |
+
"step": 3500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.09027132982564738,
|
| 331 |
+
"eval_accuracy": 0.9212555776224498,
|
| 332 |
+
"eval_f1": 0.9212388888583593,
|
| 333 |
+
"eval_loss": 0.217011496424675,
|
| 334 |
+
"eval_precision": 0.9220491698721403,
|
| 335 |
+
"eval_recall": 0.9212555776224498,
|
| 336 |
+
"eval_runtime": 82.2451,
|
| 337 |
+
"eval_samples_per_second": 942.816,
|
| 338 |
+
"eval_steps_per_second": 58.934,
|
| 339 |
+
"step": 3500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.09285051067780872,
|
| 343 |
+
"grad_norm": 3.4589858055114746,
|
| 344 |
+
"learning_rate": 1.9464840781929957e-05,
|
| 345 |
+
"loss": 0.2335,
|
| 346 |
+
"step": 3600
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.09542969152997008,
|
| 350 |
+
"grad_norm": 6.5932111740112305,
|
| 351 |
+
"learning_rate": 1.9447572010775716e-05,
|
| 352 |
+
"loss": 0.2239,
|
| 353 |
+
"step": 3700
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.09800887238213143,
|
| 357 |
+
"grad_norm": 2.6831107139587402,
|
| 358 |
+
"learning_rate": 1.943030323962147e-05,
|
| 359 |
+
"loss": 0.2049,
|
| 360 |
+
"step": 3800
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.10058805323429279,
|
| 364 |
+
"grad_norm": 26.725677490234375,
|
| 365 |
+
"learning_rate": 1.9413034468467223e-05,
|
| 366 |
+
"loss": 0.2404,
|
| 367 |
+
"step": 3900
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.10316723408645415,
|
| 371 |
+
"grad_norm": 26.966718673706055,
|
| 372 |
+
"learning_rate": 1.939576569731298e-05,
|
| 373 |
+
"loss": 0.241,
|
| 374 |
+
"step": 4000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.10316723408645415,
|
| 378 |
+
"eval_accuracy": 0.9104227386448634,
|
| 379 |
+
"eval_f1": 0.9102303033501772,
|
| 380 |
+
"eval_loss": 0.22579680383205414,
|
| 381 |
+
"eval_precision": 0.9150459252068764,
|
| 382 |
+
"eval_recall": 0.9104227386448634,
|
| 383 |
+
"eval_runtime": 81.0753,
|
| 384 |
+
"eval_samples_per_second": 956.419,
|
| 385 |
+
"eval_steps_per_second": 59.784,
|
| 386 |
+
"step": 4000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.1057464149386155,
|
| 390 |
+
"grad_norm": 17.290939331054688,
|
| 391 |
+
"learning_rate": 1.9378496926158737e-05,
|
| 392 |
+
"loss": 0.2298,
|
| 393 |
+
"step": 4100
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.10832559579077684,
|
| 397 |
+
"grad_norm": 8.844673156738281,
|
| 398 |
+
"learning_rate": 1.9361228155004492e-05,
|
| 399 |
+
"loss": 0.224,
|
| 400 |
+
"step": 4200
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.1109047766429382,
|
| 404 |
+
"grad_norm": 2.0837674140930176,
|
| 405 |
+
"learning_rate": 1.9343959383850247e-05,
|
| 406 |
+
"loss": 0.2367,
|
| 407 |
+
"step": 4300
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.11348395749509955,
|
| 411 |
+
"grad_norm": 7.639578819274902,
|
| 412 |
+
"learning_rate": 1.9326690612696002e-05,
|
| 413 |
+
"loss": 0.2194,
|
| 414 |
+
"step": 4400
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.11606313834726091,
|
| 418 |
+
"grad_norm": 9.39929485321045,
|
| 419 |
+
"learning_rate": 1.9309421841541757e-05,
|
| 420 |
+
"loss": 0.2297,
|
| 421 |
+
"step": 4500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.11606313834726091,
|
| 425 |
+
"eval_accuracy": 0.9185473678780531,
|
| 426 |
+
"eval_f1": 0.918488362019573,
|
| 427 |
+
"eval_loss": 0.24766957759857178,
|
| 428 |
+
"eval_precision": 0.9204437423697496,
|
| 429 |
+
"eval_recall": 0.9185473678780531,
|
| 430 |
+
"eval_runtime": 80.6546,
|
| 431 |
+
"eval_samples_per_second": 961.408,
|
| 432 |
+
"eval_steps_per_second": 60.096,
|
| 433 |
+
"step": 4500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.11864231919942227,
|
| 437 |
+
"grad_norm": 63.414676666259766,
|
| 438 |
+
"learning_rate": 1.9292153070387512e-05,
|
| 439 |
+
"loss": 0.2067,
|
| 440 |
+
"step": 4600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.12122150005158362,
|
| 444 |
+
"grad_norm": 27.895687103271484,
|
| 445 |
+
"learning_rate": 1.9274884299233268e-05,
|
| 446 |
+
"loss": 0.1975,
|
| 447 |
+
"step": 4700
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.12380068090374496,
|
| 451 |
+
"grad_norm": 5.536584377288818,
|
| 452 |
+
"learning_rate": 1.9257615528079023e-05,
|
| 453 |
+
"loss": 0.2141,
|
| 454 |
+
"step": 4800
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.12637986175590632,
|
| 458 |
+
"grad_norm": 18.6892147064209,
|
| 459 |
+
"learning_rate": 1.9240346756924778e-05,
|
| 460 |
+
"loss": 0.2549,
|
| 461 |
+
"step": 4900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.12895904260806768,
|
| 465 |
+
"grad_norm": 3.705787420272827,
|
| 466 |
+
"learning_rate": 1.9223077985770536e-05,
|
| 467 |
+
"loss": 0.2187,
|
| 468 |
+
"step": 5000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.12895904260806768,
|
| 472 |
+
"eval_accuracy": 0.9244409481313353,
|
| 473 |
+
"eval_f1": 0.9243951823390982,
|
| 474 |
+
"eval_loss": 0.22254011034965515,
|
| 475 |
+
"eval_precision": 0.9250643288957023,
|
| 476 |
+
"eval_recall": 0.9244409481313353,
|
| 477 |
+
"eval_runtime": 84.6134,
|
| 478 |
+
"eval_samples_per_second": 916.427,
|
| 479 |
+
"eval_steps_per_second": 57.284,
|
| 480 |
+
"step": 5000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.13153822346022903,
|
| 484 |
+
"grad_norm": 10.738173484802246,
|
| 485 |
+
"learning_rate": 1.9205809214616288e-05,
|
| 486 |
+
"loss": 0.2143,
|
| 487 |
+
"step": 5100
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.1341174043123904,
|
| 491 |
+
"grad_norm": 7.39179801940918,
|
| 492 |
+
"learning_rate": 1.9188540443462043e-05,
|
| 493 |
+
"loss": 0.205,
|
| 494 |
+
"step": 5200
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.13669658516455174,
|
| 498 |
+
"grad_norm": 14.402508735656738,
|
| 499 |
+
"learning_rate": 1.9171271672307802e-05,
|
| 500 |
+
"loss": 0.2027,
|
| 501 |
+
"step": 5300
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.1392757660167131,
|
| 505 |
+
"grad_norm": 8.077102661132812,
|
| 506 |
+
"learning_rate": 1.9154002901153554e-05,
|
| 507 |
+
"loss": 0.1909,
|
| 508 |
+
"step": 5400
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.14185494686887445,
|
| 512 |
+
"grad_norm": 2.036444664001465,
|
| 513 |
+
"learning_rate": 1.913673412999931e-05,
|
| 514 |
+
"loss": 0.2285,
|
| 515 |
+
"step": 5500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.14185494686887445,
|
| 519 |
+
"eval_accuracy": 0.9220164555982564,
|
| 520 |
+
"eval_f1": 0.9220066809051832,
|
| 521 |
+
"eval_loss": 0.2105225920677185,
|
| 522 |
+
"eval_precision": 0.9220853946451525,
|
| 523 |
+
"eval_recall": 0.9220164555982564,
|
| 524 |
+
"eval_runtime": 85.408,
|
| 525 |
+
"eval_samples_per_second": 907.901,
|
| 526 |
+
"eval_steps_per_second": 56.751,
|
| 527 |
+
"step": 5500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1444341277210358,
|
| 531 |
+
"grad_norm": 3.172539472579956,
|
| 532 |
+
"learning_rate": 1.9119465358845067e-05,
|
| 533 |
+
"loss": 0.1839,
|
| 534 |
+
"step": 5600
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.14701330857319717,
|
| 538 |
+
"grad_norm": 4.246190547943115,
|
| 539 |
+
"learning_rate": 1.9102196587690823e-05,
|
| 540 |
+
"loss": 0.2248,
|
| 541 |
+
"step": 5700
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1495924894253585,
|
| 545 |
+
"grad_norm": 9.676543235778809,
|
| 546 |
+
"learning_rate": 1.9084927816536574e-05,
|
| 547 |
+
"loss": 0.2003,
|
| 548 |
+
"step": 5800
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.15217167027751985,
|
| 552 |
+
"grad_norm": 14.093317031860352,
|
| 553 |
+
"learning_rate": 1.9067659045382333e-05,
|
| 554 |
+
"loss": 0.2283,
|
| 555 |
+
"step": 5900
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.1547508511296812,
|
| 559 |
+
"grad_norm": 6.108678817749023,
|
| 560 |
+
"learning_rate": 1.9050390274228088e-05,
|
| 561 |
+
"loss": 0.2001,
|
| 562 |
+
"step": 6000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.1547508511296812,
|
| 566 |
+
"eval_accuracy": 0.9288514611436383,
|
| 567 |
+
"eval_f1": 0.9288285810713337,
|
| 568 |
+
"eval_loss": 0.19547687470912933,
|
| 569 |
+
"eval_precision": 0.9291294558172835,
|
| 570 |
+
"eval_recall": 0.9288514611436383,
|
| 571 |
+
"eval_runtime": 85.1832,
|
| 572 |
+
"eval_samples_per_second": 910.297,
|
| 573 |
+
"eval_steps_per_second": 56.901,
|
| 574 |
+
"step": 6000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.15733003198184256,
|
| 578 |
+
"grad_norm": 25.980791091918945,
|
| 579 |
+
"learning_rate": 1.9033121503073843e-05,
|
| 580 |
+
"loss": 0.2045,
|
| 581 |
+
"step": 6100
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.15990921283400392,
|
| 585 |
+
"grad_norm": 3.697967767715454,
|
| 586 |
+
"learning_rate": 1.90158527319196e-05,
|
| 587 |
+
"loss": 0.1921,
|
| 588 |
+
"step": 6200
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.16248839368616527,
|
| 592 |
+
"grad_norm": 6.734536647796631,
|
| 593 |
+
"learning_rate": 1.8998583960765353e-05,
|
| 594 |
+
"loss": 0.2398,
|
| 595 |
+
"step": 6300
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.16506757453832663,
|
| 599 |
+
"grad_norm": 4.171863555908203,
|
| 600 |
+
"learning_rate": 1.898131518961111e-05,
|
| 601 |
+
"loss": 0.1851,
|
| 602 |
+
"step": 6400
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.16764675539048798,
|
| 606 |
+
"grad_norm": 8.22290325164795,
|
| 607 |
+
"learning_rate": 1.8964046418456864e-05,
|
| 608 |
+
"loss": 0.2048,
|
| 609 |
+
"step": 6500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.16764675539048798,
|
| 613 |
+
"eval_accuracy": 0.9237316551030409,
|
| 614 |
+
"eval_f1": 0.9237108101584136,
|
| 615 |
+
"eval_loss": 0.21942158043384552,
|
| 616 |
+
"eval_precision": 0.9246712025171661,
|
| 617 |
+
"eval_recall": 0.9237316551030409,
|
| 618 |
+
"eval_runtime": 87.4195,
|
| 619 |
+
"eval_samples_per_second": 887.01,
|
| 620 |
+
"eval_steps_per_second": 55.445,
|
| 621 |
+
"step": 6500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.17022593624264934,
|
| 625 |
+
"grad_norm": 1.4064490795135498,
|
| 626 |
+
"learning_rate": 1.894677764730262e-05,
|
| 627 |
+
"loss": 0.2192,
|
| 628 |
+
"step": 6600
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.1728051170948107,
|
| 632 |
+
"grad_norm": 0.8423879146575928,
|
| 633 |
+
"learning_rate": 1.8929508876148374e-05,
|
| 634 |
+
"loss": 0.2296,
|
| 635 |
+
"step": 6700
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.17538429794697205,
|
| 639 |
+
"grad_norm": 2.1388964653015137,
|
| 640 |
+
"learning_rate": 1.891224010499413e-05,
|
| 641 |
+
"loss": 0.2397,
|
| 642 |
+
"step": 6800
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.1779634787991334,
|
| 646 |
+
"grad_norm": 6.909358501434326,
|
| 647 |
+
"learning_rate": 1.8894971333839888e-05,
|
| 648 |
+
"loss": 0.1902,
|
| 649 |
+
"step": 6900
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.18054265965129476,
|
| 653 |
+
"grad_norm": 1.044767141342163,
|
| 654 |
+
"learning_rate": 1.887770256268564e-05,
|
| 655 |
+
"loss": 0.1931,
|
| 656 |
+
"step": 7000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.18054265965129476,
|
| 660 |
+
"eval_accuracy": 0.9318046993887183,
|
| 661 |
+
"eval_f1": 0.9318026741133476,
|
| 662 |
+
"eval_loss": 0.18849702179431915,
|
| 663 |
+
"eval_precision": 0.9318098201446341,
|
| 664 |
+
"eval_recall": 0.9318046993887183,
|
| 665 |
+
"eval_runtime": 85.7797,
|
| 666 |
+
"eval_samples_per_second": 903.967,
|
| 667 |
+
"eval_steps_per_second": 56.505,
|
| 668 |
+
"step": 7000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.1831218405034561,
|
| 672 |
+
"grad_norm": 9.8370943069458,
|
| 673 |
+
"learning_rate": 1.8860433791531395e-05,
|
| 674 |
+
"loss": 0.2179,
|
| 675 |
+
"step": 7100
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.18570102135561745,
|
| 679 |
+
"grad_norm": 3.937364101409912,
|
| 680 |
+
"learning_rate": 1.8843165020377153e-05,
|
| 681 |
+
"loss": 0.2017,
|
| 682 |
+
"step": 7200
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.1882802022077788,
|
| 686 |
+
"grad_norm": 1.7348313331604004,
|
| 687 |
+
"learning_rate": 1.8825896249222905e-05,
|
| 688 |
+
"loss": 0.2098,
|
| 689 |
+
"step": 7300
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.19085938305994016,
|
| 693 |
+
"grad_norm": 13.738706588745117,
|
| 694 |
+
"learning_rate": 1.880862747806866e-05,
|
| 695 |
+
"loss": 0.2176,
|
| 696 |
+
"step": 7400
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.1934385639121015,
|
| 700 |
+
"grad_norm": 7.888184547424316,
|
| 701 |
+
"learning_rate": 1.879135870691442e-05,
|
| 702 |
+
"loss": 0.2052,
|
| 703 |
+
"step": 7500
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.1934385639121015,
|
| 707 |
+
"eval_accuracy": 0.9288256686698821,
|
| 708 |
+
"eval_f1": 0.9287744301894623,
|
| 709 |
+
"eval_loss": 0.19062775373458862,
|
| 710 |
+
"eval_precision": 0.9296090951021123,
|
| 711 |
+
"eval_recall": 0.9288256686698821,
|
| 712 |
+
"eval_runtime": 84.5191,
|
| 713 |
+
"eval_samples_per_second": 917.45,
|
| 714 |
+
"eval_steps_per_second": 57.348,
|
| 715 |
+
"step": 7500
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.19601774476426287,
|
| 719 |
+
"grad_norm": 8.46711254119873,
|
| 720 |
+
"learning_rate": 1.8774089935760174e-05,
|
| 721 |
+
"loss": 0.2069,
|
| 722 |
+
"step": 7600
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.19859692561642422,
|
| 726 |
+
"grad_norm": 0.5938565731048584,
|
| 727 |
+
"learning_rate": 1.8756821164605926e-05,
|
| 728 |
+
"loss": 0.1842,
|
| 729 |
+
"step": 7700
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.20117610646858558,
|
| 733 |
+
"grad_norm": 14.665600776672363,
|
| 734 |
+
"learning_rate": 1.8739552393451684e-05,
|
| 735 |
+
"loss": 0.2097,
|
| 736 |
+
"step": 7800
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.20375528732074694,
|
| 740 |
+
"grad_norm": 11.763516426086426,
|
| 741 |
+
"learning_rate": 1.872228362229744e-05,
|
| 742 |
+
"loss": 0.221,
|
| 743 |
+
"step": 7900
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.2063344681729083,
|
| 747 |
+
"grad_norm": 3.6280648708343506,
|
| 748 |
+
"learning_rate": 1.8705014851143195e-05,
|
| 749 |
+
"loss": 0.1904,
|
| 750 |
+
"step": 8000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.2063344681729083,
|
| 754 |
+
"eval_accuracy": 0.929612339119445,
|
| 755 |
+
"eval_f1": 0.9296143948889304,
|
| 756 |
+
"eval_loss": 0.18757739663124084,
|
| 757 |
+
"eval_precision": 0.9297072870129078,
|
| 758 |
+
"eval_recall": 0.929612339119445,
|
| 759 |
+
"eval_runtime": 85.1458,
|
| 760 |
+
"eval_samples_per_second": 910.696,
|
| 761 |
+
"eval_steps_per_second": 56.926,
|
| 762 |
+
"step": 8000
|
| 763 |
+
}
|
| 764 |
+
],
|
| 765 |
+
"logging_steps": 100,
|
| 766 |
+
"max_steps": 116316,
|
| 767 |
+
"num_input_tokens_seen": 0,
|
| 768 |
+
"num_train_epochs": 3,
|
| 769 |
+
"save_steps": 500,
|
| 770 |
+
"stateful_callbacks": {
|
| 771 |
+
"EarlyStoppingCallback": {
|
| 772 |
+
"args": {
|
| 773 |
+
"early_stopping_patience": 3,
|
| 774 |
+
"early_stopping_threshold": 0.0
|
| 775 |
+
},
|
| 776 |
+
"attributes": {
|
| 777 |
+
"early_stopping_patience_counter": 2
|
| 778 |
+
}
|
| 779 |
+
},
|
| 780 |
+
"TrainerControl": {
|
| 781 |
+
"args": {
|
| 782 |
+
"should_epoch_stop": false,
|
| 783 |
+
"should_evaluate": false,
|
| 784 |
+
"should_log": false,
|
| 785 |
+
"should_save": true,
|
| 786 |
+
"should_training_stop": false
|
| 787 |
+
},
|
| 788 |
+
"attributes": {}
|
| 789 |
+
}
|
| 790 |
+
},
|
| 791 |
+
"total_flos": 2.894541453336672e+16,
|
| 792 |
+
"train_batch_size": 16,
|
| 793 |
+
"trial_name": null,
|
| 794 |
+
"trial_params": null
|
| 795 |
+
}
|
checkpoint-8000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce8abd265ced130b3a60da4969f0ccef6e504e9f0ee20e35110ba851d1b32cb3
|
| 3 |
+
size 5777
|
checkpoint-8500/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"XLMRobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "single_label_classification",
|
| 24 |
+
"transformers_version": "4.57.3",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
checkpoint-8500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:630287452f619f9b63d672a2de963e2387815a61e169ce3e8c4e38eaf3443f8c
|
| 3 |
+
size 1112205008
|
checkpoint-8500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76fb241df89cbf9a7ef9352c71ea051668d3f466ab6b31ddfe1d6affd2eed6ab
|
| 3 |
+
size 2224532875
|
checkpoint-8500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69835862f1cc5fe7a491a3efa7ffa0138b3b8637580ba7909e3f8c41ee34228a
|
| 3 |
+
size 14645
|
checkpoint-8500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6ed9cd0c6034218e5c6ffcb7beab99f8695dfbf2fff5bfc25f34cdf042e26d2
|
| 3 |
+
size 1383
|
checkpoint-8500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6280d3b52579c204059f19675b839e105c8667272a05af43b8e228df65db087f
|
| 3 |
+
size 1465
|
checkpoint-8500/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-8500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-8500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "<mask>",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"pad_token": "<pad>",
|
| 52 |
+
"sep_token": "</s>",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
checkpoint-8500/trainer_state.json
ADDED
|
@@ -0,0 +1,842 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 7000,
|
| 3 |
+
"best_metric": 0.9318026741133476,
|
| 4 |
+
"best_model_checkpoint": "./results/checkpoint-7000",
|
| 5 |
+
"epoch": 0.21923037243371504,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 8500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0025791808521613536,
|
| 14 |
+
"grad_norm": 8.55178451538086,
|
| 15 |
+
"learning_rate": 3.96e-06,
|
| 16 |
+
"loss": 0.6948,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.005158361704322707,
|
| 21 |
+
"grad_norm": 6.15341329574585,
|
| 22 |
+
"learning_rate": 7.960000000000002e-06,
|
| 23 |
+
"loss": 0.6256,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00773754255648406,
|
| 28 |
+
"grad_norm": 7.876453876495361,
|
| 29 |
+
"learning_rate": 1.196e-05,
|
| 30 |
+
"loss": 0.4996,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.010316723408645414,
|
| 35 |
+
"grad_norm": 25.46845817565918,
|
| 36 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 37 |
+
"loss": 0.4267,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.012895904260806768,
|
| 42 |
+
"grad_norm": 12.137995719909668,
|
| 43 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 44 |
+
"loss": 0.3932,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.012895904260806768,
|
| 49 |
+
"eval_accuracy": 0.8497975290810141,
|
| 50 |
+
"eval_f1": 0.8492073290894184,
|
| 51 |
+
"eval_loss": 0.36694180965423584,
|
| 52 |
+
"eval_precision": 0.8566847895515343,
|
| 53 |
+
"eval_recall": 0.8497975290810141,
|
| 54 |
+
"eval_runtime": 81.3998,
|
| 55 |
+
"eval_samples_per_second": 952.607,
|
| 56 |
+
"eval_steps_per_second": 59.546,
|
| 57 |
+
"step": 500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.01547508511296812,
|
| 61 |
+
"grad_norm": 9.444395065307617,
|
| 62 |
+
"learning_rate": 1.99829039165573e-05,
|
| 63 |
+
"loss": 0.3438,
|
| 64 |
+
"step": 600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.018054265965129476,
|
| 68 |
+
"grad_norm": 12.698025703430176,
|
| 69 |
+
"learning_rate": 1.9965635145403055e-05,
|
| 70 |
+
"loss": 0.3897,
|
| 71 |
+
"step": 700
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.02063344681729083,
|
| 75 |
+
"grad_norm": 7.648699760437012,
|
| 76 |
+
"learning_rate": 1.994836637424881e-05,
|
| 77 |
+
"loss": 0.3197,
|
| 78 |
+
"step": 800
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.02321262766945218,
|
| 82 |
+
"grad_norm": 11.770590782165527,
|
| 83 |
+
"learning_rate": 1.9931097603094565e-05,
|
| 84 |
+
"loss": 0.2641,
|
| 85 |
+
"step": 900
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.025791808521613536,
|
| 89 |
+
"grad_norm": 1.2174805402755737,
|
| 90 |
+
"learning_rate": 1.991382883194032e-05,
|
| 91 |
+
"loss": 0.3093,
|
| 92 |
+
"step": 1000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.025791808521613536,
|
| 96 |
+
"eval_accuracy": 0.8805163653245983,
|
| 97 |
+
"eval_f1": 0.8798423409034712,
|
| 98 |
+
"eval_loss": 0.29810118675231934,
|
| 99 |
+
"eval_precision": 0.8908826043373803,
|
| 100 |
+
"eval_recall": 0.8805163653245983,
|
| 101 |
+
"eval_runtime": 82.3661,
|
| 102 |
+
"eval_samples_per_second": 941.431,
|
| 103 |
+
"eval_steps_per_second": 58.847,
|
| 104 |
+
"step": 1000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.02837098937377489,
|
| 108 |
+
"grad_norm": 35.91494369506836,
|
| 109 |
+
"learning_rate": 1.9896560060786075e-05,
|
| 110 |
+
"loss": 0.2701,
|
| 111 |
+
"step": 1100
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.03095017022593624,
|
| 115 |
+
"grad_norm": 10.724688529968262,
|
| 116 |
+
"learning_rate": 1.9879291289631834e-05,
|
| 117 |
+
"loss": 0.2723,
|
| 118 |
+
"step": 1200
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.0335293510780976,
|
| 122 |
+
"grad_norm": 11.826756477355957,
|
| 123 |
+
"learning_rate": 1.9862022518477585e-05,
|
| 124 |
+
"loss": 0.2669,
|
| 125 |
+
"step": 1300
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.03610853193025895,
|
| 129 |
+
"grad_norm": 13.958097457885742,
|
| 130 |
+
"learning_rate": 1.984475374732334e-05,
|
| 131 |
+
"loss": 0.2563,
|
| 132 |
+
"step": 1400
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.0386877127824203,
|
| 136 |
+
"grad_norm": 32.344303131103516,
|
| 137 |
+
"learning_rate": 1.98274849761691e-05,
|
| 138 |
+
"loss": 0.2717,
|
| 139 |
+
"step": 1500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.0386877127824203,
|
| 143 |
+
"eval_accuracy": 0.9098424079853499,
|
| 144 |
+
"eval_f1": 0.9098439407901873,
|
| 145 |
+
"eval_loss": 0.2351153939962387,
|
| 146 |
+
"eval_precision": 0.9100045564108226,
|
| 147 |
+
"eval_recall": 0.9098424079853499,
|
| 148 |
+
"eval_runtime": 80.9696,
|
| 149 |
+
"eval_samples_per_second": 957.669,
|
| 150 |
+
"eval_steps_per_second": 59.862,
|
| 151 |
+
"step": 1500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.04126689363458166,
|
| 155 |
+
"grad_norm": 41.87895584106445,
|
| 156 |
+
"learning_rate": 1.9810216205014854e-05,
|
| 157 |
+
"loss": 0.2775,
|
| 158 |
+
"step": 1600
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.04384607448674301,
|
| 162 |
+
"grad_norm": 9.029143333435059,
|
| 163 |
+
"learning_rate": 1.9792947433860606e-05,
|
| 164 |
+
"loss": 0.2747,
|
| 165 |
+
"step": 1700
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.04642525533890436,
|
| 169 |
+
"grad_norm": 0.5453509092330933,
|
| 170 |
+
"learning_rate": 1.9775678662706365e-05,
|
| 171 |
+
"loss": 0.2333,
|
| 172 |
+
"step": 1800
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.04900443619106572,
|
| 176 |
+
"grad_norm": 14.877126693725586,
|
| 177 |
+
"learning_rate": 1.975840989155212e-05,
|
| 178 |
+
"loss": 0.2335,
|
| 179 |
+
"step": 1900
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.05158361704322707,
|
| 183 |
+
"grad_norm": 9.152324676513672,
|
| 184 |
+
"learning_rate": 1.974114112039787e-05,
|
| 185 |
+
"loss": 0.2501,
|
| 186 |
+
"step": 2000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.05158361704322707,
|
| 190 |
+
"eval_accuracy": 0.915400686079802,
|
| 191 |
+
"eval_f1": 0.9153786857019529,
|
| 192 |
+
"eval_loss": 0.24714471399784088,
|
| 193 |
+
"eval_precision": 0.9155981210913413,
|
| 194 |
+
"eval_recall": 0.915400686079802,
|
| 195 |
+
"eval_runtime": 80.7839,
|
| 196 |
+
"eval_samples_per_second": 959.869,
|
| 197 |
+
"eval_steps_per_second": 60.0,
|
| 198 |
+
"step": 2000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.05416279789538842,
|
| 202 |
+
"grad_norm": 3.9494690895080566,
|
| 203 |
+
"learning_rate": 1.972387234924363e-05,
|
| 204 |
+
"loss": 0.2432,
|
| 205 |
+
"step": 2100
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.05674197874754978,
|
| 209 |
+
"grad_norm": 7.518190383911133,
|
| 210 |
+
"learning_rate": 1.9706603578089385e-05,
|
| 211 |
+
"loss": 0.2179,
|
| 212 |
+
"step": 2200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.05932115959971113,
|
| 216 |
+
"grad_norm": 55.069461822509766,
|
| 217 |
+
"learning_rate": 1.968933480693514e-05,
|
| 218 |
+
"loss": 0.2603,
|
| 219 |
+
"step": 2300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.06190034045187248,
|
| 223 |
+
"grad_norm": 21.235607147216797,
|
| 224 |
+
"learning_rate": 1.9672066035780896e-05,
|
| 225 |
+
"loss": 0.2363,
|
| 226 |
+
"step": 2400
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.06447952130403384,
|
| 230 |
+
"grad_norm": 30.956710815429688,
|
| 231 |
+
"learning_rate": 1.965479726462665e-05,
|
| 232 |
+
"loss": 0.264,
|
| 233 |
+
"step": 2500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06447952130403384,
|
| 237 |
+
"eval_accuracy": 0.9174769802171726,
|
| 238 |
+
"eval_f1": 0.9174732815674826,
|
| 239 |
+
"eval_loss": 0.22409705817699432,
|
| 240 |
+
"eval_precision": 0.9178450257324412,
|
| 241 |
+
"eval_recall": 0.9174769802171726,
|
| 242 |
+
"eval_runtime": 82.4939,
|
| 243 |
+
"eval_samples_per_second": 939.972,
|
| 244 |
+
"eval_steps_per_second": 58.756,
|
| 245 |
+
"step": 2500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.0670587021561952,
|
| 249 |
+
"grad_norm": 3.0290513038635254,
|
| 250 |
+
"learning_rate": 1.9637528493472406e-05,
|
| 251 |
+
"loss": 0.2388,
|
| 252 |
+
"step": 2600
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.06963788300835655,
|
| 256 |
+
"grad_norm": 4.662931442260742,
|
| 257 |
+
"learning_rate": 1.962025972231816e-05,
|
| 258 |
+
"loss": 0.2132,
|
| 259 |
+
"step": 2700
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.0722170638605179,
|
| 263 |
+
"grad_norm": 31.94866180419922,
|
| 264 |
+
"learning_rate": 1.9602990951163916e-05,
|
| 265 |
+
"loss": 0.2576,
|
| 266 |
+
"step": 2800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.07479624471267925,
|
| 270 |
+
"grad_norm": 3.5957560539245605,
|
| 271 |
+
"learning_rate": 1.958572218000967e-05,
|
| 272 |
+
"loss": 0.2404,
|
| 273 |
+
"step": 2900
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.0773754255648406,
|
| 277 |
+
"grad_norm": 6.48631477355957,
|
| 278 |
+
"learning_rate": 1.9568453408855427e-05,
|
| 279 |
+
"loss": 0.2514,
|
| 280 |
+
"step": 3000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.0773754255648406,
|
| 284 |
+
"eval_accuracy": 0.9169998194526837,
|
| 285 |
+
"eval_f1": 0.9170017211060641,
|
| 286 |
+
"eval_loss": 0.21086551249027252,
|
| 287 |
+
"eval_precision": 0.9171348873594787,
|
| 288 |
+
"eval_recall": 0.9169998194526837,
|
| 289 |
+
"eval_runtime": 80.1992,
|
| 290 |
+
"eval_samples_per_second": 966.867,
|
| 291 |
+
"eval_steps_per_second": 60.437,
|
| 292 |
+
"step": 3000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.07995460641700196,
|
| 296 |
+
"grad_norm": 5.789911270141602,
|
| 297 |
+
"learning_rate": 1.9551184637701185e-05,
|
| 298 |
+
"loss": 0.2138,
|
| 299 |
+
"step": 3100
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.08253378726916331,
|
| 303 |
+
"grad_norm": 8.613458633422852,
|
| 304 |
+
"learning_rate": 1.9533915866546937e-05,
|
| 305 |
+
"loss": 0.2323,
|
| 306 |
+
"step": 3200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.08511296812132467,
|
| 310 |
+
"grad_norm": 5.5360822677612305,
|
| 311 |
+
"learning_rate": 1.9516647095392692e-05,
|
| 312 |
+
"loss": 0.2474,
|
| 313 |
+
"step": 3300
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08769214897348603,
|
| 317 |
+
"grad_norm": 15.3199462890625,
|
| 318 |
+
"learning_rate": 1.949937832423845e-05,
|
| 319 |
+
"loss": 0.1964,
|
| 320 |
+
"step": 3400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.09027132982564738,
|
| 324 |
+
"grad_norm": 0.23756052553653717,
|
| 325 |
+
"learning_rate": 1.9482109553084202e-05,
|
| 326 |
+
"loss": 0.2048,
|
| 327 |
+
"step": 3500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.09027132982564738,
|
| 331 |
+
"eval_accuracy": 0.9212555776224498,
|
| 332 |
+
"eval_f1": 0.9212388888583593,
|
| 333 |
+
"eval_loss": 0.217011496424675,
|
| 334 |
+
"eval_precision": 0.9220491698721403,
|
| 335 |
+
"eval_recall": 0.9212555776224498,
|
| 336 |
+
"eval_runtime": 82.2451,
|
| 337 |
+
"eval_samples_per_second": 942.816,
|
| 338 |
+
"eval_steps_per_second": 58.934,
|
| 339 |
+
"step": 3500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.09285051067780872,
|
| 343 |
+
"grad_norm": 3.4589858055114746,
|
| 344 |
+
"learning_rate": 1.9464840781929957e-05,
|
| 345 |
+
"loss": 0.2335,
|
| 346 |
+
"step": 3600
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.09542969152997008,
|
| 350 |
+
"grad_norm": 6.5932111740112305,
|
| 351 |
+
"learning_rate": 1.9447572010775716e-05,
|
| 352 |
+
"loss": 0.2239,
|
| 353 |
+
"step": 3700
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.09800887238213143,
|
| 357 |
+
"grad_norm": 2.6831107139587402,
|
| 358 |
+
"learning_rate": 1.943030323962147e-05,
|
| 359 |
+
"loss": 0.2049,
|
| 360 |
+
"step": 3800
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.10058805323429279,
|
| 364 |
+
"grad_norm": 26.725677490234375,
|
| 365 |
+
"learning_rate": 1.9413034468467223e-05,
|
| 366 |
+
"loss": 0.2404,
|
| 367 |
+
"step": 3900
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.10316723408645415,
|
| 371 |
+
"grad_norm": 26.966718673706055,
|
| 372 |
+
"learning_rate": 1.939576569731298e-05,
|
| 373 |
+
"loss": 0.241,
|
| 374 |
+
"step": 4000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.10316723408645415,
|
| 378 |
+
"eval_accuracy": 0.9104227386448634,
|
| 379 |
+
"eval_f1": 0.9102303033501772,
|
| 380 |
+
"eval_loss": 0.22579680383205414,
|
| 381 |
+
"eval_precision": 0.9150459252068764,
|
| 382 |
+
"eval_recall": 0.9104227386448634,
|
| 383 |
+
"eval_runtime": 81.0753,
|
| 384 |
+
"eval_samples_per_second": 956.419,
|
| 385 |
+
"eval_steps_per_second": 59.784,
|
| 386 |
+
"step": 4000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.1057464149386155,
|
| 390 |
+
"grad_norm": 17.290939331054688,
|
| 391 |
+
"learning_rate": 1.9378496926158737e-05,
|
| 392 |
+
"loss": 0.2298,
|
| 393 |
+
"step": 4100
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.10832559579077684,
|
| 397 |
+
"grad_norm": 8.844673156738281,
|
| 398 |
+
"learning_rate": 1.9361228155004492e-05,
|
| 399 |
+
"loss": 0.224,
|
| 400 |
+
"step": 4200
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.1109047766429382,
|
| 404 |
+
"grad_norm": 2.0837674140930176,
|
| 405 |
+
"learning_rate": 1.9343959383850247e-05,
|
| 406 |
+
"loss": 0.2367,
|
| 407 |
+
"step": 4300
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.11348395749509955,
|
| 411 |
+
"grad_norm": 7.639578819274902,
|
| 412 |
+
"learning_rate": 1.9326690612696002e-05,
|
| 413 |
+
"loss": 0.2194,
|
| 414 |
+
"step": 4400
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.11606313834726091,
|
| 418 |
+
"grad_norm": 9.39929485321045,
|
| 419 |
+
"learning_rate": 1.9309421841541757e-05,
|
| 420 |
+
"loss": 0.2297,
|
| 421 |
+
"step": 4500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.11606313834726091,
|
| 425 |
+
"eval_accuracy": 0.9185473678780531,
|
| 426 |
+
"eval_f1": 0.918488362019573,
|
| 427 |
+
"eval_loss": 0.24766957759857178,
|
| 428 |
+
"eval_precision": 0.9204437423697496,
|
| 429 |
+
"eval_recall": 0.9185473678780531,
|
| 430 |
+
"eval_runtime": 80.6546,
|
| 431 |
+
"eval_samples_per_second": 961.408,
|
| 432 |
+
"eval_steps_per_second": 60.096,
|
| 433 |
+
"step": 4500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.11864231919942227,
|
| 437 |
+
"grad_norm": 63.414676666259766,
|
| 438 |
+
"learning_rate": 1.9292153070387512e-05,
|
| 439 |
+
"loss": 0.2067,
|
| 440 |
+
"step": 4600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.12122150005158362,
|
| 444 |
+
"grad_norm": 27.895687103271484,
|
| 445 |
+
"learning_rate": 1.9274884299233268e-05,
|
| 446 |
+
"loss": 0.1975,
|
| 447 |
+
"step": 4700
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.12380068090374496,
|
| 451 |
+
"grad_norm": 5.536584377288818,
|
| 452 |
+
"learning_rate": 1.9257615528079023e-05,
|
| 453 |
+
"loss": 0.2141,
|
| 454 |
+
"step": 4800
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.12637986175590632,
|
| 458 |
+
"grad_norm": 18.6892147064209,
|
| 459 |
+
"learning_rate": 1.9240346756924778e-05,
|
| 460 |
+
"loss": 0.2549,
|
| 461 |
+
"step": 4900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.12895904260806768,
|
| 465 |
+
"grad_norm": 3.705787420272827,
|
| 466 |
+
"learning_rate": 1.9223077985770536e-05,
|
| 467 |
+
"loss": 0.2187,
|
| 468 |
+
"step": 5000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.12895904260806768,
|
| 472 |
+
"eval_accuracy": 0.9244409481313353,
|
| 473 |
+
"eval_f1": 0.9243951823390982,
|
| 474 |
+
"eval_loss": 0.22254011034965515,
|
| 475 |
+
"eval_precision": 0.9250643288957023,
|
| 476 |
+
"eval_recall": 0.9244409481313353,
|
| 477 |
+
"eval_runtime": 84.6134,
|
| 478 |
+
"eval_samples_per_second": 916.427,
|
| 479 |
+
"eval_steps_per_second": 57.284,
|
| 480 |
+
"step": 5000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.13153822346022903,
|
| 484 |
+
"grad_norm": 10.738173484802246,
|
| 485 |
+
"learning_rate": 1.9205809214616288e-05,
|
| 486 |
+
"loss": 0.2143,
|
| 487 |
+
"step": 5100
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.1341174043123904,
|
| 491 |
+
"grad_norm": 7.39179801940918,
|
| 492 |
+
"learning_rate": 1.9188540443462043e-05,
|
| 493 |
+
"loss": 0.205,
|
| 494 |
+
"step": 5200
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.13669658516455174,
|
| 498 |
+
"grad_norm": 14.402508735656738,
|
| 499 |
+
"learning_rate": 1.9171271672307802e-05,
|
| 500 |
+
"loss": 0.2027,
|
| 501 |
+
"step": 5300
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.1392757660167131,
|
| 505 |
+
"grad_norm": 8.077102661132812,
|
| 506 |
+
"learning_rate": 1.9154002901153554e-05,
|
| 507 |
+
"loss": 0.1909,
|
| 508 |
+
"step": 5400
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.14185494686887445,
|
| 512 |
+
"grad_norm": 2.036444664001465,
|
| 513 |
+
"learning_rate": 1.913673412999931e-05,
|
| 514 |
+
"loss": 0.2285,
|
| 515 |
+
"step": 5500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.14185494686887445,
|
| 519 |
+
"eval_accuracy": 0.9220164555982564,
|
| 520 |
+
"eval_f1": 0.9220066809051832,
|
| 521 |
+
"eval_loss": 0.2105225920677185,
|
| 522 |
+
"eval_precision": 0.9220853946451525,
|
| 523 |
+
"eval_recall": 0.9220164555982564,
|
| 524 |
+
"eval_runtime": 85.408,
|
| 525 |
+
"eval_samples_per_second": 907.901,
|
| 526 |
+
"eval_steps_per_second": 56.751,
|
| 527 |
+
"step": 5500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1444341277210358,
|
| 531 |
+
"grad_norm": 3.172539472579956,
|
| 532 |
+
"learning_rate": 1.9119465358845067e-05,
|
| 533 |
+
"loss": 0.1839,
|
| 534 |
+
"step": 5600
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.14701330857319717,
|
| 538 |
+
"grad_norm": 4.246190547943115,
|
| 539 |
+
"learning_rate": 1.9102196587690823e-05,
|
| 540 |
+
"loss": 0.2248,
|
| 541 |
+
"step": 5700
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1495924894253585,
|
| 545 |
+
"grad_norm": 9.676543235778809,
|
| 546 |
+
"learning_rate": 1.9084927816536574e-05,
|
| 547 |
+
"loss": 0.2003,
|
| 548 |
+
"step": 5800
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.15217167027751985,
|
| 552 |
+
"grad_norm": 14.093317031860352,
|
| 553 |
+
"learning_rate": 1.9067659045382333e-05,
|
| 554 |
+
"loss": 0.2283,
|
| 555 |
+
"step": 5900
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.1547508511296812,
|
| 559 |
+
"grad_norm": 6.108678817749023,
|
| 560 |
+
"learning_rate": 1.9050390274228088e-05,
|
| 561 |
+
"loss": 0.2001,
|
| 562 |
+
"step": 6000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.1547508511296812,
|
| 566 |
+
"eval_accuracy": 0.9288514611436383,
|
| 567 |
+
"eval_f1": 0.9288285810713337,
|
| 568 |
+
"eval_loss": 0.19547687470912933,
|
| 569 |
+
"eval_precision": 0.9291294558172835,
|
| 570 |
+
"eval_recall": 0.9288514611436383,
|
| 571 |
+
"eval_runtime": 85.1832,
|
| 572 |
+
"eval_samples_per_second": 910.297,
|
| 573 |
+
"eval_steps_per_second": 56.901,
|
| 574 |
+
"step": 6000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.15733003198184256,
|
| 578 |
+
"grad_norm": 25.980791091918945,
|
| 579 |
+
"learning_rate": 1.9033121503073843e-05,
|
| 580 |
+
"loss": 0.2045,
|
| 581 |
+
"step": 6100
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.15990921283400392,
|
| 585 |
+
"grad_norm": 3.697967767715454,
|
| 586 |
+
"learning_rate": 1.90158527319196e-05,
|
| 587 |
+
"loss": 0.1921,
|
| 588 |
+
"step": 6200
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.16248839368616527,
|
| 592 |
+
"grad_norm": 6.734536647796631,
|
| 593 |
+
"learning_rate": 1.8998583960765353e-05,
|
| 594 |
+
"loss": 0.2398,
|
| 595 |
+
"step": 6300
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.16506757453832663,
|
| 599 |
+
"grad_norm": 4.171863555908203,
|
| 600 |
+
"learning_rate": 1.898131518961111e-05,
|
| 601 |
+
"loss": 0.1851,
|
| 602 |
+
"step": 6400
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.16764675539048798,
|
| 606 |
+
"grad_norm": 8.22290325164795,
|
| 607 |
+
"learning_rate": 1.8964046418456864e-05,
|
| 608 |
+
"loss": 0.2048,
|
| 609 |
+
"step": 6500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.16764675539048798,
|
| 613 |
+
"eval_accuracy": 0.9237316551030409,
|
| 614 |
+
"eval_f1": 0.9237108101584136,
|
| 615 |
+
"eval_loss": 0.21942158043384552,
|
| 616 |
+
"eval_precision": 0.9246712025171661,
|
| 617 |
+
"eval_recall": 0.9237316551030409,
|
| 618 |
+
"eval_runtime": 87.4195,
|
| 619 |
+
"eval_samples_per_second": 887.01,
|
| 620 |
+
"eval_steps_per_second": 55.445,
|
| 621 |
+
"step": 6500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.17022593624264934,
|
| 625 |
+
"grad_norm": 1.4064490795135498,
|
| 626 |
+
"learning_rate": 1.894677764730262e-05,
|
| 627 |
+
"loss": 0.2192,
|
| 628 |
+
"step": 6600
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.1728051170948107,
|
| 632 |
+
"grad_norm": 0.8423879146575928,
|
| 633 |
+
"learning_rate": 1.8929508876148374e-05,
|
| 634 |
+
"loss": 0.2296,
|
| 635 |
+
"step": 6700
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.17538429794697205,
|
| 639 |
+
"grad_norm": 2.1388964653015137,
|
| 640 |
+
"learning_rate": 1.891224010499413e-05,
|
| 641 |
+
"loss": 0.2397,
|
| 642 |
+
"step": 6800
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.1779634787991334,
|
| 646 |
+
"grad_norm": 6.909358501434326,
|
| 647 |
+
"learning_rate": 1.8894971333839888e-05,
|
| 648 |
+
"loss": 0.1902,
|
| 649 |
+
"step": 6900
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.18054265965129476,
|
| 653 |
+
"grad_norm": 1.044767141342163,
|
| 654 |
+
"learning_rate": 1.887770256268564e-05,
|
| 655 |
+
"loss": 0.1931,
|
| 656 |
+
"step": 7000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.18054265965129476,
|
| 660 |
+
"eval_accuracy": 0.9318046993887183,
|
| 661 |
+
"eval_f1": 0.9318026741133476,
|
| 662 |
+
"eval_loss": 0.18849702179431915,
|
| 663 |
+
"eval_precision": 0.9318098201446341,
|
| 664 |
+
"eval_recall": 0.9318046993887183,
|
| 665 |
+
"eval_runtime": 85.7797,
|
| 666 |
+
"eval_samples_per_second": 903.967,
|
| 667 |
+
"eval_steps_per_second": 56.505,
|
| 668 |
+
"step": 7000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.1831218405034561,
|
| 672 |
+
"grad_norm": 9.8370943069458,
|
| 673 |
+
"learning_rate": 1.8860433791531395e-05,
|
| 674 |
+
"loss": 0.2179,
|
| 675 |
+
"step": 7100
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.18570102135561745,
|
| 679 |
+
"grad_norm": 3.937364101409912,
|
| 680 |
+
"learning_rate": 1.8843165020377153e-05,
|
| 681 |
+
"loss": 0.2017,
|
| 682 |
+
"step": 7200
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.1882802022077788,
|
| 686 |
+
"grad_norm": 1.7348313331604004,
|
| 687 |
+
"learning_rate": 1.8825896249222905e-05,
|
| 688 |
+
"loss": 0.2098,
|
| 689 |
+
"step": 7300
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.19085938305994016,
|
| 693 |
+
"grad_norm": 13.738706588745117,
|
| 694 |
+
"learning_rate": 1.880862747806866e-05,
|
| 695 |
+
"loss": 0.2176,
|
| 696 |
+
"step": 7400
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.1934385639121015,
|
| 700 |
+
"grad_norm": 7.888184547424316,
|
| 701 |
+
"learning_rate": 1.879135870691442e-05,
|
| 702 |
+
"loss": 0.2052,
|
| 703 |
+
"step": 7500
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.1934385639121015,
|
| 707 |
+
"eval_accuracy": 0.9288256686698821,
|
| 708 |
+
"eval_f1": 0.9287744301894623,
|
| 709 |
+
"eval_loss": 0.19062775373458862,
|
| 710 |
+
"eval_precision": 0.9296090951021123,
|
| 711 |
+
"eval_recall": 0.9288256686698821,
|
| 712 |
+
"eval_runtime": 84.5191,
|
| 713 |
+
"eval_samples_per_second": 917.45,
|
| 714 |
+
"eval_steps_per_second": 57.348,
|
| 715 |
+
"step": 7500
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.19601774476426287,
|
| 719 |
+
"grad_norm": 8.46711254119873,
|
| 720 |
+
"learning_rate": 1.8774089935760174e-05,
|
| 721 |
+
"loss": 0.2069,
|
| 722 |
+
"step": 7600
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.19859692561642422,
|
| 726 |
+
"grad_norm": 0.5938565731048584,
|
| 727 |
+
"learning_rate": 1.8756821164605926e-05,
|
| 728 |
+
"loss": 0.1842,
|
| 729 |
+
"step": 7700
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.20117610646858558,
|
| 733 |
+
"grad_norm": 14.665600776672363,
|
| 734 |
+
"learning_rate": 1.8739552393451684e-05,
|
| 735 |
+
"loss": 0.2097,
|
| 736 |
+
"step": 7800
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.20375528732074694,
|
| 740 |
+
"grad_norm": 11.763516426086426,
|
| 741 |
+
"learning_rate": 1.872228362229744e-05,
|
| 742 |
+
"loss": 0.221,
|
| 743 |
+
"step": 7900
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.2063344681729083,
|
| 747 |
+
"grad_norm": 3.6280648708343506,
|
| 748 |
+
"learning_rate": 1.8705014851143195e-05,
|
| 749 |
+
"loss": 0.1904,
|
| 750 |
+
"step": 8000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.2063344681729083,
|
| 754 |
+
"eval_accuracy": 0.929612339119445,
|
| 755 |
+
"eval_f1": 0.9296143948889304,
|
| 756 |
+
"eval_loss": 0.18757739663124084,
|
| 757 |
+
"eval_precision": 0.9297072870129078,
|
| 758 |
+
"eval_recall": 0.929612339119445,
|
| 759 |
+
"eval_runtime": 85.1458,
|
| 760 |
+
"eval_samples_per_second": 910.696,
|
| 761 |
+
"eval_steps_per_second": 56.926,
|
| 762 |
+
"step": 8000
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.20891364902506965,
|
| 766 |
+
"grad_norm": 15.164484977722168,
|
| 767 |
+
"learning_rate": 1.868774607998895e-05,
|
| 768 |
+
"loss": 0.1958,
|
| 769 |
+
"step": 8100
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.211492829877231,
|
| 773 |
+
"grad_norm": 2.8772385120391846,
|
| 774 |
+
"learning_rate": 1.8670477308834705e-05,
|
| 775 |
+
"loss": 0.1878,
|
| 776 |
+
"step": 8200
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.21407201072939233,
|
| 780 |
+
"grad_norm": 2.487945079803467,
|
| 781 |
+
"learning_rate": 1.865320853768046e-05,
|
| 782 |
+
"loss": 0.2033,
|
| 783 |
+
"step": 8300
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.2166511915815537,
|
| 787 |
+
"grad_norm": 9.773316383361816,
|
| 788 |
+
"learning_rate": 1.8635939766526215e-05,
|
| 789 |
+
"loss": 0.1709,
|
| 790 |
+
"step": 8400
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.21923037243371504,
|
| 794 |
+
"grad_norm": 11.347258567810059,
|
| 795 |
+
"learning_rate": 1.861867099537197e-05,
|
| 796 |
+
"loss": 0.2125,
|
| 797 |
+
"step": 8500
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.21923037243371504,
|
| 801 |
+
"eval_accuracy": 0.9292125557762245,
|
| 802 |
+
"eval_f1": 0.9291884106571301,
|
| 803 |
+
"eval_loss": 0.18593531847000122,
|
| 804 |
+
"eval_precision": 0.9303160172840454,
|
| 805 |
+
"eval_recall": 0.9292125557762245,
|
| 806 |
+
"eval_runtime": 82.8305,
|
| 807 |
+
"eval_samples_per_second": 936.153,
|
| 808 |
+
"eval_steps_per_second": 58.517,
|
| 809 |
+
"step": 8500
|
| 810 |
+
}
|
| 811 |
+
],
|
| 812 |
+
"logging_steps": 100,
|
| 813 |
+
"max_steps": 116316,
|
| 814 |
+
"num_input_tokens_seen": 0,
|
| 815 |
+
"num_train_epochs": 3,
|
| 816 |
+
"save_steps": 500,
|
| 817 |
+
"stateful_callbacks": {
|
| 818 |
+
"EarlyStoppingCallback": {
|
| 819 |
+
"args": {
|
| 820 |
+
"early_stopping_patience": 3,
|
| 821 |
+
"early_stopping_threshold": 0.0
|
| 822 |
+
},
|
| 823 |
+
"attributes": {
|
| 824 |
+
"early_stopping_patience_counter": 3
|
| 825 |
+
}
|
| 826 |
+
},
|
| 827 |
+
"TrainerControl": {
|
| 828 |
+
"args": {
|
| 829 |
+
"should_epoch_stop": false,
|
| 830 |
+
"should_evaluate": false,
|
| 831 |
+
"should_log": false,
|
| 832 |
+
"should_save": true,
|
| 833 |
+
"should_training_stop": true
|
| 834 |
+
},
|
| 835 |
+
"attributes": {}
|
| 836 |
+
}
|
| 837 |
+
},
|
| 838 |
+
"total_flos": 3.072344504550528e+16,
|
| 839 |
+
"train_batch_size": 16,
|
| 840 |
+
"trial_name": null,
|
| 841 |
+
"trial_params": null
|
| 842 |
+
}
|
checkpoint-8500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce8abd265ced130b3a60da4969f0ccef6e504e9f0ee20e35110ba851d1b32cb3
|
| 3 |
+
size 5777
|
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"XLMRobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 12,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "single_label_classification",
|
| 24 |
+
"transformers_version": "4.57.3",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11581fcf59ed6effc51225813116b13565a436317e8381e00444ac3ce13d695e
|
| 3 |
+
size 1112205008
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
test.py
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 2 |
+
import torch
|
| 3 |
+
|
| 4 |
+
MODEL_PATH = "model.safetensors" # папка с model.safetensors + config.json
|
| 5 |
+
|
| 6 |
+
tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
|
| 7 |
+
model = AutoModelForSequenceClassification.from_pretrained(MODEL_PATH)
|
| 8 |
+
|
| 9 |
+
def predict(text: str):
|
| 10 |
+
inputs = tokenizer(
|
| 11 |
+
text,
|
| 12 |
+
return_tensors="pt",
|
| 13 |
+
truncation=True,
|
| 14 |
+
padding=True,
|
| 15 |
+
max_length=512
|
| 16 |
+
)
|
| 17 |
+
|
| 18 |
+
with torch.no_grad():
|
| 19 |
+
outputs = model(**inputs)
|
| 20 |
+
|
| 21 |
+
logits = outputs.logits
|
| 22 |
+
probs = torch.softmax(logits, dim=-1)
|
| 23 |
+
|
| 24 |
+
pred_id = torch.argmax(probs, dim=-1).item()
|
| 25 |
+
confidence = probs[0, pred_id].item()
|
| 26 |
+
|
| 27 |
+
return {
|
| 28 |
+
"label_id": pred_id,
|
| 29 |
+
"confidence": confidence
|
| 30 |
+
}
|
| 31 |
+
|
| 32 |
+
print(predict("Салем калын калай?"))
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": false,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "<mask>",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"pad_token": "<pad>",
|
| 52 |
+
"sep_token": "</s>",
|
| 53 |
+
"sp_model_kwargs": {},
|
| 54 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 55 |
+
"unk_token": "<unk>"
|
| 56 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce8abd265ced130b3a60da4969f0ccef6e504e9f0ee20e35110ba851d1b32cb3
|
| 3 |
+
size 5777
|