Upload training checkpoints and best model
Browse files- checkpoint-1000/config.json +29 -0
- checkpoint-1000/merges.txt +0 -0
- checkpoint-1000/model.safetensors +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +51 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer_config.json +58 -0
- checkpoint-1000/trainer_state.json +769 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-1000/vocab.json +0 -0
- checkpoint-1500/config.json +29 -0
- checkpoint-1500/merges.txt +0 -0
- checkpoint-1500/model.safetensors +3 -0
- checkpoint-1500/optimizer.pt +3 -0
- checkpoint-1500/rng_state.pth +3 -0
- checkpoint-1500/scheduler.pt +3 -0
- checkpoint-1500/special_tokens_map.json +51 -0
- checkpoint-1500/tokenizer.json +0 -0
- checkpoint-1500/tokenizer_config.json +58 -0
- checkpoint-1500/trainer_state.json +1132 -0
- checkpoint-1500/training_args.bin +3 -0
- checkpoint-1500/vocab.json +0 -0
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 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 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 1026,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"problem_type": "single_label_classification",
|
| 25 |
+
"transformers_version": "4.56.0",
|
| 26 |
+
"type_vocab_size": 10,
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 51416
|
| 29 |
+
}
|
checkpoint-1000/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d51378b54e35c00d17c3ac2006f9e9fdbd7cc12e52e1278758b6df39d27a87a
|
| 3 |
+
size 503749208
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cce9d35e3077c44ed259aba54f709f632f5bf67a59fa794ed03bf3be1f17490a
|
| 3 |
+
size 1007621899
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9681354556c442279f8d15fdd4995c6180c0aa2fc961cc72bad8bdd7491d3e7
|
| 3 |
+
size 14645
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ca762945532cf70776f01dc50c979557aef924ea20456b955387653adde5a87
|
| 3 |
+
size 1977
|
checkpoint-1000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-1000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,769 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 0.9613078781032101,
|
| 4 |
+
"best_model_checkpoint": "taskA-unixcoder-focal/checkpoint-1000",
|
| 5 |
+
"epoch": 0.064,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 3.3732223510742188,
|
| 15 |
+
"learning_rate": 9.6e-08,
|
| 16 |
+
"loss": 0.2065,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 2.8028833866119385,
|
| 22 |
+
"learning_rate": 2.0266666666666668e-07,
|
| 23 |
+
"loss": 0.2003,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 3.137962818145752,
|
| 29 |
+
"learning_rate": 3.0933333333333335e-07,
|
| 30 |
+
"loss": 0.2119,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 2.2841057777404785,
|
| 36 |
+
"learning_rate": 4.16e-07,
|
| 37 |
+
"loss": 0.1858,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 2.4040842056274414,
|
| 43 |
+
"learning_rate": 5.226666666666667e-07,
|
| 44 |
+
"loss": 0.1755,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 2.1078808307647705,
|
| 50 |
+
"learning_rate": 6.293333333333334e-07,
|
| 51 |
+
"loss": 0.1836,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 2.1723551750183105,
|
| 57 |
+
"learning_rate": 7.36e-07,
|
| 58 |
+
"loss": 0.182,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 3.8697516918182373,
|
| 64 |
+
"learning_rate": 8.426666666666668e-07,
|
| 65 |
+
"loss": 0.1893,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 2.2711849212646484,
|
| 71 |
+
"learning_rate": 9.493333333333334e-07,
|
| 72 |
+
"loss": 0.1645,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 2.9138245582580566,
|
| 78 |
+
"learning_rate": 1.0560000000000001e-06,
|
| 79 |
+
"loss": 0.1753,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 2.136301279067993,
|
| 85 |
+
"learning_rate": 1.1626666666666667e-06,
|
| 86 |
+
"loss": 0.1639,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 3.216343641281128,
|
| 92 |
+
"learning_rate": 1.2693333333333335e-06,
|
| 93 |
+
"loss": 0.1623,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 3.192094326019287,
|
| 99 |
+
"learning_rate": 1.376e-06,
|
| 100 |
+
"loss": 0.1607,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 4.722238540649414,
|
| 106 |
+
"learning_rate": 1.4826666666666666e-06,
|
| 107 |
+
"loss": 0.1653,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 4.088972568511963,
|
| 113 |
+
"learning_rate": 1.5893333333333336e-06,
|
| 114 |
+
"loss": 0.1557,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.2773237228393555,
|
| 120 |
+
"learning_rate": 1.6960000000000002e-06,
|
| 121 |
+
"loss": 0.1634,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 2.53987193107605,
|
| 127 |
+
"learning_rate": 1.8026666666666667e-06,
|
| 128 |
+
"loss": 0.1435,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 3.688382387161255,
|
| 134 |
+
"learning_rate": 1.9093333333333335e-06,
|
| 135 |
+
"loss": 0.1396,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.8855538368225098,
|
| 141 |
+
"learning_rate": 2.0160000000000003e-06,
|
| 142 |
+
"loss": 0.1476,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 2.2272040843963623,
|
| 148 |
+
"learning_rate": 2.1226666666666666e-06,
|
| 149 |
+
"loss": 0.144,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 3.505934476852417,
|
| 155 |
+
"learning_rate": 2.229333333333334e-06,
|
| 156 |
+
"loss": 0.1306,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.8632491827011108,
|
| 162 |
+
"learning_rate": 2.336e-06,
|
| 163 |
+
"loss": 0.1504,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 2.191713333129883,
|
| 169 |
+
"learning_rate": 2.442666666666667e-06,
|
| 170 |
+
"loss": 0.1346,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 1.9042261838912964,
|
| 176 |
+
"learning_rate": 2.5493333333333337e-06,
|
| 177 |
+
"loss": 0.135,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 2.266148090362549,
|
| 183 |
+
"learning_rate": 2.656e-06,
|
| 184 |
+
"loss": 0.1356,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 2.4776418209075928,
|
| 190 |
+
"learning_rate": 2.762666666666667e-06,
|
| 191 |
+
"loss": 0.129,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 2.340693712234497,
|
| 197 |
+
"learning_rate": 2.8693333333333336e-06,
|
| 198 |
+
"loss": 0.1236,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6849262714385986,
|
| 204 |
+
"learning_rate": 2.976e-06,
|
| 205 |
+
"loss": 0.1152,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 3.5535154342651367,
|
| 211 |
+
"learning_rate": 3.082666666666667e-06,
|
| 212 |
+
"loss": 0.1335,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 2.9307031631469727,
|
| 218 |
+
"learning_rate": 3.1893333333333335e-06,
|
| 219 |
+
"loss": 0.1136,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 3.0100226402282715,
|
| 225 |
+
"learning_rate": 3.2960000000000003e-06,
|
| 226 |
+
"loss": 0.1186,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 2.088911771774292,
|
| 232 |
+
"learning_rate": 3.402666666666667e-06,
|
| 233 |
+
"loss": 0.1051,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 4.332556247711182,
|
| 239 |
+
"learning_rate": 3.5093333333333334e-06,
|
| 240 |
+
"loss": 0.1075,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 2.954036235809326,
|
| 246 |
+
"learning_rate": 3.616e-06,
|
| 247 |
+
"loss": 0.11,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.262308359146118,
|
| 253 |
+
"learning_rate": 3.7226666666666674e-06,
|
| 254 |
+
"loss": 0.1057,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.5145989656448364,
|
| 260 |
+
"learning_rate": 3.829333333333334e-06,
|
| 261 |
+
"loss": 0.1049,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.740663766860962,
|
| 267 |
+
"learning_rate": 3.936e-06,
|
| 268 |
+
"loss": 0.0964,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.7438851594924927,
|
| 274 |
+
"learning_rate": 4.042666666666667e-06,
|
| 275 |
+
"loss": 0.0936,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 3.946115493774414,
|
| 281 |
+
"learning_rate": 4.149333333333334e-06,
|
| 282 |
+
"loss": 0.1038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 5.289428234100342,
|
| 288 |
+
"learning_rate": 4.256e-06,
|
| 289 |
+
"loss": 0.1015,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 1.9344863891601562,
|
| 295 |
+
"learning_rate": 4.362666666666667e-06,
|
| 296 |
+
"loss": 0.0855,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 7.482412338256836,
|
| 302 |
+
"learning_rate": 4.469333333333334e-06,
|
| 303 |
+
"loss": 0.1031,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 2.5646328926086426,
|
| 309 |
+
"learning_rate": 4.576000000000001e-06,
|
| 310 |
+
"loss": 0.0777,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.473496437072754,
|
| 316 |
+
"learning_rate": 4.682666666666667e-06,
|
| 317 |
+
"loss": 0.0854,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 2.204341411590576,
|
| 323 |
+
"learning_rate": 4.7893333333333334e-06,
|
| 324 |
+
"loss": 0.0739,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 3.1363987922668457,
|
| 330 |
+
"learning_rate": 4.896e-06,
|
| 331 |
+
"loss": 0.0886,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 2.0473029613494873,
|
| 337 |
+
"learning_rate": 5.002666666666667e-06,
|
| 338 |
+
"loss": 0.0794,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.6088207960128784,
|
| 344 |
+
"learning_rate": 5.109333333333334e-06,
|
| 345 |
+
"loss": 0.0604,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 2.974717617034912,
|
| 351 |
+
"learning_rate": 5.216e-06,
|
| 352 |
+
"loss": 0.0827,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 6.114630699157715,
|
| 358 |
+
"learning_rate": 5.322666666666667e-06,
|
| 359 |
+
"loss": 0.0631,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.032,
|
| 364 |
+
"eval_accuracy": 0.88603,
|
| 365 |
+
"eval_f1_weighted": 0.8857328199768308,
|
| 366 |
+
"eval_loss": 0.08366190642118454,
|
| 367 |
+
"eval_macro_f1": 0.8859061279979327,
|
| 368 |
+
"eval_precision": 0.8966428429175579,
|
| 369 |
+
"eval_recall": 0.88603,
|
| 370 |
+
"eval_runtime": 1629.6546,
|
| 371 |
+
"eval_samples_per_second": 61.363,
|
| 372 |
+
"eval_steps_per_second": 0.959,
|
| 373 |
+
"step": 500
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.03264,
|
| 377 |
+
"grad_norm": 1.4639947414398193,
|
| 378 |
+
"learning_rate": 5.429333333333334e-06,
|
| 379 |
+
"loss": 0.0724,
|
| 380 |
+
"step": 510
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.03328,
|
| 384 |
+
"grad_norm": 4.234114646911621,
|
| 385 |
+
"learning_rate": 5.536e-06,
|
| 386 |
+
"loss": 0.0737,
|
| 387 |
+
"step": 520
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.03392,
|
| 391 |
+
"grad_norm": 1.9391822814941406,
|
| 392 |
+
"learning_rate": 5.642666666666668e-06,
|
| 393 |
+
"loss": 0.0875,
|
| 394 |
+
"step": 530
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.03456,
|
| 398 |
+
"grad_norm": 2.0388095378875732,
|
| 399 |
+
"learning_rate": 5.7493333333333336e-06,
|
| 400 |
+
"loss": 0.0661,
|
| 401 |
+
"step": 540
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.0352,
|
| 405 |
+
"grad_norm": 1.9697074890136719,
|
| 406 |
+
"learning_rate": 5.856e-06,
|
| 407 |
+
"loss": 0.0647,
|
| 408 |
+
"step": 550
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.03584,
|
| 412 |
+
"grad_norm": 3.9906229972839355,
|
| 413 |
+
"learning_rate": 5.962666666666666e-06,
|
| 414 |
+
"loss": 0.0696,
|
| 415 |
+
"step": 560
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.03648,
|
| 419 |
+
"grad_norm": 1.5516153573989868,
|
| 420 |
+
"learning_rate": 6.069333333333334e-06,
|
| 421 |
+
"loss": 0.0576,
|
| 422 |
+
"step": 570
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.03712,
|
| 426 |
+
"grad_norm": 3.5216825008392334,
|
| 427 |
+
"learning_rate": 6.176000000000001e-06,
|
| 428 |
+
"loss": 0.0793,
|
| 429 |
+
"step": 580
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.03776,
|
| 433 |
+
"grad_norm": 6.22791051864624,
|
| 434 |
+
"learning_rate": 6.282666666666667e-06,
|
| 435 |
+
"loss": 0.0757,
|
| 436 |
+
"step": 590
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.0384,
|
| 440 |
+
"grad_norm": 1.8255369663238525,
|
| 441 |
+
"learning_rate": 6.389333333333334e-06,
|
| 442 |
+
"loss": 0.0602,
|
| 443 |
+
"step": 600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.03904,
|
| 447 |
+
"grad_norm": 2.506188154220581,
|
| 448 |
+
"learning_rate": 6.496e-06,
|
| 449 |
+
"loss": 0.0643,
|
| 450 |
+
"step": 610
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.03968,
|
| 454 |
+
"grad_norm": 1.5877528190612793,
|
| 455 |
+
"learning_rate": 6.602666666666667e-06,
|
| 456 |
+
"loss": 0.0501,
|
| 457 |
+
"step": 620
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.04032,
|
| 461 |
+
"grad_norm": 2.1468632221221924,
|
| 462 |
+
"learning_rate": 6.7093333333333345e-06,
|
| 463 |
+
"loss": 0.0521,
|
| 464 |
+
"step": 630
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.04096,
|
| 468 |
+
"grad_norm": 2.6225597858428955,
|
| 469 |
+
"learning_rate": 6.8160000000000005e-06,
|
| 470 |
+
"loss": 0.068,
|
| 471 |
+
"step": 640
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.0416,
|
| 475 |
+
"grad_norm": 3.1171772480010986,
|
| 476 |
+
"learning_rate": 6.922666666666667e-06,
|
| 477 |
+
"loss": 0.0644,
|
| 478 |
+
"step": 650
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.04224,
|
| 482 |
+
"grad_norm": 2.7589240074157715,
|
| 483 |
+
"learning_rate": 7.029333333333333e-06,
|
| 484 |
+
"loss": 0.0763,
|
| 485 |
+
"step": 660
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.04288,
|
| 489 |
+
"grad_norm": 2.3155124187469482,
|
| 490 |
+
"learning_rate": 7.136000000000001e-06,
|
| 491 |
+
"loss": 0.0451,
|
| 492 |
+
"step": 670
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.04352,
|
| 496 |
+
"grad_norm": 0.9795209169387817,
|
| 497 |
+
"learning_rate": 7.242666666666667e-06,
|
| 498 |
+
"loss": 0.0552,
|
| 499 |
+
"step": 680
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.04416,
|
| 503 |
+
"grad_norm": 6.224135398864746,
|
| 504 |
+
"learning_rate": 7.3493333333333335e-06,
|
| 505 |
+
"loss": 0.0614,
|
| 506 |
+
"step": 690
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.0448,
|
| 510 |
+
"grad_norm": 1.9188823699951172,
|
| 511 |
+
"learning_rate": 7.456000000000001e-06,
|
| 512 |
+
"loss": 0.0539,
|
| 513 |
+
"step": 700
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.04544,
|
| 517 |
+
"grad_norm": 2.0834364891052246,
|
| 518 |
+
"learning_rate": 7.562666666666667e-06,
|
| 519 |
+
"loss": 0.0545,
|
| 520 |
+
"step": 710
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.04608,
|
| 524 |
+
"grad_norm": 2.081609010696411,
|
| 525 |
+
"learning_rate": 7.669333333333335e-06,
|
| 526 |
+
"loss": 0.0563,
|
| 527 |
+
"step": 720
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.04672,
|
| 531 |
+
"grad_norm": 1.461228370666504,
|
| 532 |
+
"learning_rate": 7.776e-06,
|
| 533 |
+
"loss": 0.0495,
|
| 534 |
+
"step": 730
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.04736,
|
| 538 |
+
"grad_norm": 2.7746951580047607,
|
| 539 |
+
"learning_rate": 7.882666666666667e-06,
|
| 540 |
+
"loss": 0.0517,
|
| 541 |
+
"step": 740
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.048,
|
| 545 |
+
"grad_norm": 1.2801299095153809,
|
| 546 |
+
"learning_rate": 7.989333333333335e-06,
|
| 547 |
+
"loss": 0.034,
|
| 548 |
+
"step": 750
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.04864,
|
| 552 |
+
"grad_norm": 3.921491861343384,
|
| 553 |
+
"learning_rate": 8.096e-06,
|
| 554 |
+
"loss": 0.0537,
|
| 555 |
+
"step": 760
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.04928,
|
| 559 |
+
"grad_norm": 1.9986109733581543,
|
| 560 |
+
"learning_rate": 8.202666666666667e-06,
|
| 561 |
+
"loss": 0.0444,
|
| 562 |
+
"step": 770
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.04992,
|
| 566 |
+
"grad_norm": 4.678001403808594,
|
| 567 |
+
"learning_rate": 8.309333333333334e-06,
|
| 568 |
+
"loss": 0.0589,
|
| 569 |
+
"step": 780
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.05056,
|
| 573 |
+
"grad_norm": 1.7302327156066895,
|
| 574 |
+
"learning_rate": 8.416e-06,
|
| 575 |
+
"loss": 0.043,
|
| 576 |
+
"step": 790
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.0512,
|
| 580 |
+
"grad_norm": 2.5643601417541504,
|
| 581 |
+
"learning_rate": 8.522666666666667e-06,
|
| 582 |
+
"loss": 0.0671,
|
| 583 |
+
"step": 800
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.05184,
|
| 587 |
+
"grad_norm": 2.8869082927703857,
|
| 588 |
+
"learning_rate": 8.629333333333334e-06,
|
| 589 |
+
"loss": 0.0565,
|
| 590 |
+
"step": 810
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.05248,
|
| 594 |
+
"grad_norm": 3.723088264465332,
|
| 595 |
+
"learning_rate": 8.736e-06,
|
| 596 |
+
"loss": 0.0428,
|
| 597 |
+
"step": 820
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.05312,
|
| 601 |
+
"grad_norm": 4.387404441833496,
|
| 602 |
+
"learning_rate": 8.842666666666667e-06,
|
| 603 |
+
"loss": 0.0559,
|
| 604 |
+
"step": 830
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.05376,
|
| 608 |
+
"grad_norm": 1.889413833618164,
|
| 609 |
+
"learning_rate": 8.949333333333334e-06,
|
| 610 |
+
"loss": 0.0405,
|
| 611 |
+
"step": 840
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.0544,
|
| 615 |
+
"grad_norm": 1.4773402214050293,
|
| 616 |
+
"learning_rate": 9.056000000000001e-06,
|
| 617 |
+
"loss": 0.0414,
|
| 618 |
+
"step": 850
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.05504,
|
| 622 |
+
"grad_norm": 0.4495583176612854,
|
| 623 |
+
"learning_rate": 9.162666666666668e-06,
|
| 624 |
+
"loss": 0.0446,
|
| 625 |
+
"step": 860
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.05568,
|
| 629 |
+
"grad_norm": 1.158181071281433,
|
| 630 |
+
"learning_rate": 9.269333333333335e-06,
|
| 631 |
+
"loss": 0.0432,
|
| 632 |
+
"step": 870
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.05632,
|
| 636 |
+
"grad_norm": 2.434063196182251,
|
| 637 |
+
"learning_rate": 9.376000000000001e-06,
|
| 638 |
+
"loss": 0.0502,
|
| 639 |
+
"step": 880
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.05696,
|
| 643 |
+
"grad_norm": 1.2009385824203491,
|
| 644 |
+
"learning_rate": 9.482666666666668e-06,
|
| 645 |
+
"loss": 0.0473,
|
| 646 |
+
"step": 890
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.0576,
|
| 650 |
+
"grad_norm": 3.9361023902893066,
|
| 651 |
+
"learning_rate": 9.589333333333333e-06,
|
| 652 |
+
"loss": 0.046,
|
| 653 |
+
"step": 900
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.05824,
|
| 657 |
+
"grad_norm": 2.926988124847412,
|
| 658 |
+
"learning_rate": 9.696000000000002e-06,
|
| 659 |
+
"loss": 0.0464,
|
| 660 |
+
"step": 910
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.05888,
|
| 664 |
+
"grad_norm": 2.215726375579834,
|
| 665 |
+
"learning_rate": 9.802666666666667e-06,
|
| 666 |
+
"loss": 0.0471,
|
| 667 |
+
"step": 920
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.05952,
|
| 671 |
+
"grad_norm": 2.4104645252227783,
|
| 672 |
+
"learning_rate": 9.909333333333334e-06,
|
| 673 |
+
"loss": 0.0482,
|
| 674 |
+
"step": 930
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.06016,
|
| 678 |
+
"grad_norm": 5.664385795593262,
|
| 679 |
+
"learning_rate": 1.0016000000000002e-05,
|
| 680 |
+
"loss": 0.044,
|
| 681 |
+
"step": 940
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.0608,
|
| 685 |
+
"grad_norm": 1.7977851629257202,
|
| 686 |
+
"learning_rate": 1.0122666666666667e-05,
|
| 687 |
+
"loss": 0.0429,
|
| 688 |
+
"step": 950
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.06144,
|
| 692 |
+
"grad_norm": 2.953789472579956,
|
| 693 |
+
"learning_rate": 1.0229333333333334e-05,
|
| 694 |
+
"loss": 0.0533,
|
| 695 |
+
"step": 960
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.06208,
|
| 699 |
+
"grad_norm": 1.6272608041763306,
|
| 700 |
+
"learning_rate": 1.0336000000000002e-05,
|
| 701 |
+
"loss": 0.0307,
|
| 702 |
+
"step": 970
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.06272,
|
| 706 |
+
"grad_norm": 2.8621678352355957,
|
| 707 |
+
"learning_rate": 1.0442666666666667e-05,
|
| 708 |
+
"loss": 0.0461,
|
| 709 |
+
"step": 980
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.06336,
|
| 713 |
+
"grad_norm": 3.70019793510437,
|
| 714 |
+
"learning_rate": 1.0549333333333334e-05,
|
| 715 |
+
"loss": 0.0546,
|
| 716 |
+
"step": 990
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.064,
|
| 720 |
+
"grad_norm": 1.1777608394622803,
|
| 721 |
+
"learning_rate": 1.0656000000000003e-05,
|
| 722 |
+
"loss": 0.0508,
|
| 723 |
+
"step": 1000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.064,
|
| 727 |
+
"eval_accuracy": 0.96139,
|
| 728 |
+
"eval_f1_weighted": 0.9613900534763816,
|
| 729 |
+
"eval_loss": 0.031560152769088745,
|
| 730 |
+
"eval_macro_f1": 0.9613078781032101,
|
| 731 |
+
"eval_precision": 0.9613901086866451,
|
| 732 |
+
"eval_recall": 0.96139,
|
| 733 |
+
"eval_runtime": 1630.3285,
|
| 734 |
+
"eval_samples_per_second": 61.337,
|
| 735 |
+
"eval_steps_per_second": 0.959,
|
| 736 |
+
"step": 1000
|
| 737 |
+
}
|
| 738 |
+
],
|
| 739 |
+
"logging_steps": 10,
|
| 740 |
+
"max_steps": 15625,
|
| 741 |
+
"num_input_tokens_seen": 0,
|
| 742 |
+
"num_train_epochs": 1,
|
| 743 |
+
"save_steps": 500,
|
| 744 |
+
"stateful_callbacks": {
|
| 745 |
+
"EarlyStoppingCallback": {
|
| 746 |
+
"args": {
|
| 747 |
+
"early_stopping_patience": 3,
|
| 748 |
+
"early_stopping_threshold": 0.0
|
| 749 |
+
},
|
| 750 |
+
"attributes": {
|
| 751 |
+
"early_stopping_patience_counter": 0
|
| 752 |
+
}
|
| 753 |
+
},
|
| 754 |
+
"TrainerControl": {
|
| 755 |
+
"args": {
|
| 756 |
+
"should_epoch_stop": false,
|
| 757 |
+
"should_evaluate": false,
|
| 758 |
+
"should_log": false,
|
| 759 |
+
"should_save": true,
|
| 760 |
+
"should_training_stop": false
|
| 761 |
+
},
|
| 762 |
+
"attributes": {}
|
| 763 |
+
}
|
| 764 |
+
},
|
| 765 |
+
"total_flos": 8415459105720960.0,
|
| 766 |
+
"train_batch_size": 32,
|
| 767 |
+
"trial_name": null,
|
| 768 |
+
"trial_params": null
|
| 769 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0256e15114b0df6f165bb3e6d75f279805d6f125574ac927fca7a602f9449f4a
|
| 3 |
+
size 5777
|
checkpoint-1000/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 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 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 1026,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"problem_type": "single_label_classification",
|
| 25 |
+
"transformers_version": "4.56.0",
|
| 26 |
+
"type_vocab_size": 10,
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 51416
|
| 29 |
+
}
|
checkpoint-1500/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34ece475069d40c0d3602879a9d44cacccb53d92fdcf843008331c0076805e1a
|
| 3 |
+
size 503749208
|
checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d198b2916f97edec4db8f6e6e1de0fbe2238e1e5f87ecbe27e73cd142a9bec33
|
| 3 |
+
size 1007621899
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20c3b034723ddd19d4164d9ddced831b3a9857904fd0f88d784b134000843f0a
|
| 3 |
+
size 14645
|
checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0690ce65a5c08741822b66ea60503c9fbb2e04d3b8bd9a51620fb356f773ab32
|
| 3 |
+
size 1977
|
checkpoint-1500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-1500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,1132 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1500,
|
| 3 |
+
"best_metric": 0.9698851462608304,
|
| 4 |
+
"best_model_checkpoint": "taskA-unixcoder-focal/checkpoint-1500",
|
| 5 |
+
"epoch": 0.096,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 3.3732223510742188,
|
| 15 |
+
"learning_rate": 9.6e-08,
|
| 16 |
+
"loss": 0.2065,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 2.8028833866119385,
|
| 22 |
+
"learning_rate": 2.0266666666666668e-07,
|
| 23 |
+
"loss": 0.2003,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 3.137962818145752,
|
| 29 |
+
"learning_rate": 3.0933333333333335e-07,
|
| 30 |
+
"loss": 0.2119,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 2.2841057777404785,
|
| 36 |
+
"learning_rate": 4.16e-07,
|
| 37 |
+
"loss": 0.1858,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 2.4040842056274414,
|
| 43 |
+
"learning_rate": 5.226666666666667e-07,
|
| 44 |
+
"loss": 0.1755,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 2.1078808307647705,
|
| 50 |
+
"learning_rate": 6.293333333333334e-07,
|
| 51 |
+
"loss": 0.1836,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 2.1723551750183105,
|
| 57 |
+
"learning_rate": 7.36e-07,
|
| 58 |
+
"loss": 0.182,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 3.8697516918182373,
|
| 64 |
+
"learning_rate": 8.426666666666668e-07,
|
| 65 |
+
"loss": 0.1893,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 2.2711849212646484,
|
| 71 |
+
"learning_rate": 9.493333333333334e-07,
|
| 72 |
+
"loss": 0.1645,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 2.9138245582580566,
|
| 78 |
+
"learning_rate": 1.0560000000000001e-06,
|
| 79 |
+
"loss": 0.1753,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 2.136301279067993,
|
| 85 |
+
"learning_rate": 1.1626666666666667e-06,
|
| 86 |
+
"loss": 0.1639,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 3.216343641281128,
|
| 92 |
+
"learning_rate": 1.2693333333333335e-06,
|
| 93 |
+
"loss": 0.1623,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 3.192094326019287,
|
| 99 |
+
"learning_rate": 1.376e-06,
|
| 100 |
+
"loss": 0.1607,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 4.722238540649414,
|
| 106 |
+
"learning_rate": 1.4826666666666666e-06,
|
| 107 |
+
"loss": 0.1653,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 4.088972568511963,
|
| 113 |
+
"learning_rate": 1.5893333333333336e-06,
|
| 114 |
+
"loss": 0.1557,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.2773237228393555,
|
| 120 |
+
"learning_rate": 1.6960000000000002e-06,
|
| 121 |
+
"loss": 0.1634,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 2.53987193107605,
|
| 127 |
+
"learning_rate": 1.8026666666666667e-06,
|
| 128 |
+
"loss": 0.1435,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 3.688382387161255,
|
| 134 |
+
"learning_rate": 1.9093333333333335e-06,
|
| 135 |
+
"loss": 0.1396,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.8855538368225098,
|
| 141 |
+
"learning_rate": 2.0160000000000003e-06,
|
| 142 |
+
"loss": 0.1476,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 2.2272040843963623,
|
| 148 |
+
"learning_rate": 2.1226666666666666e-06,
|
| 149 |
+
"loss": 0.144,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 3.505934476852417,
|
| 155 |
+
"learning_rate": 2.229333333333334e-06,
|
| 156 |
+
"loss": 0.1306,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.8632491827011108,
|
| 162 |
+
"learning_rate": 2.336e-06,
|
| 163 |
+
"loss": 0.1504,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 2.191713333129883,
|
| 169 |
+
"learning_rate": 2.442666666666667e-06,
|
| 170 |
+
"loss": 0.1346,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 1.9042261838912964,
|
| 176 |
+
"learning_rate": 2.5493333333333337e-06,
|
| 177 |
+
"loss": 0.135,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 2.266148090362549,
|
| 183 |
+
"learning_rate": 2.656e-06,
|
| 184 |
+
"loss": 0.1356,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 2.4776418209075928,
|
| 190 |
+
"learning_rate": 2.762666666666667e-06,
|
| 191 |
+
"loss": 0.129,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 2.340693712234497,
|
| 197 |
+
"learning_rate": 2.8693333333333336e-06,
|
| 198 |
+
"loss": 0.1236,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6849262714385986,
|
| 204 |
+
"learning_rate": 2.976e-06,
|
| 205 |
+
"loss": 0.1152,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 3.5535154342651367,
|
| 211 |
+
"learning_rate": 3.082666666666667e-06,
|
| 212 |
+
"loss": 0.1335,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 2.9307031631469727,
|
| 218 |
+
"learning_rate": 3.1893333333333335e-06,
|
| 219 |
+
"loss": 0.1136,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 3.0100226402282715,
|
| 225 |
+
"learning_rate": 3.2960000000000003e-06,
|
| 226 |
+
"loss": 0.1186,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 2.088911771774292,
|
| 232 |
+
"learning_rate": 3.402666666666667e-06,
|
| 233 |
+
"loss": 0.1051,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 4.332556247711182,
|
| 239 |
+
"learning_rate": 3.5093333333333334e-06,
|
| 240 |
+
"loss": 0.1075,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 2.954036235809326,
|
| 246 |
+
"learning_rate": 3.616e-06,
|
| 247 |
+
"loss": 0.11,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.262308359146118,
|
| 253 |
+
"learning_rate": 3.7226666666666674e-06,
|
| 254 |
+
"loss": 0.1057,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.5145989656448364,
|
| 260 |
+
"learning_rate": 3.829333333333334e-06,
|
| 261 |
+
"loss": 0.1049,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.740663766860962,
|
| 267 |
+
"learning_rate": 3.936e-06,
|
| 268 |
+
"loss": 0.0964,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.7438851594924927,
|
| 274 |
+
"learning_rate": 4.042666666666667e-06,
|
| 275 |
+
"loss": 0.0936,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 3.946115493774414,
|
| 281 |
+
"learning_rate": 4.149333333333334e-06,
|
| 282 |
+
"loss": 0.1038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 5.289428234100342,
|
| 288 |
+
"learning_rate": 4.256e-06,
|
| 289 |
+
"loss": 0.1015,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 1.9344863891601562,
|
| 295 |
+
"learning_rate": 4.362666666666667e-06,
|
| 296 |
+
"loss": 0.0855,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 7.482412338256836,
|
| 302 |
+
"learning_rate": 4.469333333333334e-06,
|
| 303 |
+
"loss": 0.1031,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 2.5646328926086426,
|
| 309 |
+
"learning_rate": 4.576000000000001e-06,
|
| 310 |
+
"loss": 0.0777,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.473496437072754,
|
| 316 |
+
"learning_rate": 4.682666666666667e-06,
|
| 317 |
+
"loss": 0.0854,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 2.204341411590576,
|
| 323 |
+
"learning_rate": 4.7893333333333334e-06,
|
| 324 |
+
"loss": 0.0739,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 3.1363987922668457,
|
| 330 |
+
"learning_rate": 4.896e-06,
|
| 331 |
+
"loss": 0.0886,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 2.0473029613494873,
|
| 337 |
+
"learning_rate": 5.002666666666667e-06,
|
| 338 |
+
"loss": 0.0794,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.6088207960128784,
|
| 344 |
+
"learning_rate": 5.109333333333334e-06,
|
| 345 |
+
"loss": 0.0604,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 2.974717617034912,
|
| 351 |
+
"learning_rate": 5.216e-06,
|
| 352 |
+
"loss": 0.0827,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 6.114630699157715,
|
| 358 |
+
"learning_rate": 5.322666666666667e-06,
|
| 359 |
+
"loss": 0.0631,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.032,
|
| 364 |
+
"eval_accuracy": 0.88603,
|
| 365 |
+
"eval_f1_weighted": 0.8857328199768308,
|
| 366 |
+
"eval_loss": 0.08366190642118454,
|
| 367 |
+
"eval_macro_f1": 0.8859061279979327,
|
| 368 |
+
"eval_precision": 0.8966428429175579,
|
| 369 |
+
"eval_recall": 0.88603,
|
| 370 |
+
"eval_runtime": 1629.6546,
|
| 371 |
+
"eval_samples_per_second": 61.363,
|
| 372 |
+
"eval_steps_per_second": 0.959,
|
| 373 |
+
"step": 500
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.03264,
|
| 377 |
+
"grad_norm": 1.4639947414398193,
|
| 378 |
+
"learning_rate": 5.429333333333334e-06,
|
| 379 |
+
"loss": 0.0724,
|
| 380 |
+
"step": 510
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.03328,
|
| 384 |
+
"grad_norm": 4.234114646911621,
|
| 385 |
+
"learning_rate": 5.536e-06,
|
| 386 |
+
"loss": 0.0737,
|
| 387 |
+
"step": 520
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.03392,
|
| 391 |
+
"grad_norm": 1.9391822814941406,
|
| 392 |
+
"learning_rate": 5.642666666666668e-06,
|
| 393 |
+
"loss": 0.0875,
|
| 394 |
+
"step": 530
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.03456,
|
| 398 |
+
"grad_norm": 2.0388095378875732,
|
| 399 |
+
"learning_rate": 5.7493333333333336e-06,
|
| 400 |
+
"loss": 0.0661,
|
| 401 |
+
"step": 540
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.0352,
|
| 405 |
+
"grad_norm": 1.9697074890136719,
|
| 406 |
+
"learning_rate": 5.856e-06,
|
| 407 |
+
"loss": 0.0647,
|
| 408 |
+
"step": 550
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.03584,
|
| 412 |
+
"grad_norm": 3.9906229972839355,
|
| 413 |
+
"learning_rate": 5.962666666666666e-06,
|
| 414 |
+
"loss": 0.0696,
|
| 415 |
+
"step": 560
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.03648,
|
| 419 |
+
"grad_norm": 1.5516153573989868,
|
| 420 |
+
"learning_rate": 6.069333333333334e-06,
|
| 421 |
+
"loss": 0.0576,
|
| 422 |
+
"step": 570
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.03712,
|
| 426 |
+
"grad_norm": 3.5216825008392334,
|
| 427 |
+
"learning_rate": 6.176000000000001e-06,
|
| 428 |
+
"loss": 0.0793,
|
| 429 |
+
"step": 580
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.03776,
|
| 433 |
+
"grad_norm": 6.22791051864624,
|
| 434 |
+
"learning_rate": 6.282666666666667e-06,
|
| 435 |
+
"loss": 0.0757,
|
| 436 |
+
"step": 590
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.0384,
|
| 440 |
+
"grad_norm": 1.8255369663238525,
|
| 441 |
+
"learning_rate": 6.389333333333334e-06,
|
| 442 |
+
"loss": 0.0602,
|
| 443 |
+
"step": 600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.03904,
|
| 447 |
+
"grad_norm": 2.506188154220581,
|
| 448 |
+
"learning_rate": 6.496e-06,
|
| 449 |
+
"loss": 0.0643,
|
| 450 |
+
"step": 610
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.03968,
|
| 454 |
+
"grad_norm": 1.5877528190612793,
|
| 455 |
+
"learning_rate": 6.602666666666667e-06,
|
| 456 |
+
"loss": 0.0501,
|
| 457 |
+
"step": 620
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.04032,
|
| 461 |
+
"grad_norm": 2.1468632221221924,
|
| 462 |
+
"learning_rate": 6.7093333333333345e-06,
|
| 463 |
+
"loss": 0.0521,
|
| 464 |
+
"step": 630
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.04096,
|
| 468 |
+
"grad_norm": 2.6225597858428955,
|
| 469 |
+
"learning_rate": 6.8160000000000005e-06,
|
| 470 |
+
"loss": 0.068,
|
| 471 |
+
"step": 640
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.0416,
|
| 475 |
+
"grad_norm": 3.1171772480010986,
|
| 476 |
+
"learning_rate": 6.922666666666667e-06,
|
| 477 |
+
"loss": 0.0644,
|
| 478 |
+
"step": 650
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.04224,
|
| 482 |
+
"grad_norm": 2.7589240074157715,
|
| 483 |
+
"learning_rate": 7.029333333333333e-06,
|
| 484 |
+
"loss": 0.0763,
|
| 485 |
+
"step": 660
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.04288,
|
| 489 |
+
"grad_norm": 2.3155124187469482,
|
| 490 |
+
"learning_rate": 7.136000000000001e-06,
|
| 491 |
+
"loss": 0.0451,
|
| 492 |
+
"step": 670
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.04352,
|
| 496 |
+
"grad_norm": 0.9795209169387817,
|
| 497 |
+
"learning_rate": 7.242666666666667e-06,
|
| 498 |
+
"loss": 0.0552,
|
| 499 |
+
"step": 680
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.04416,
|
| 503 |
+
"grad_norm": 6.224135398864746,
|
| 504 |
+
"learning_rate": 7.3493333333333335e-06,
|
| 505 |
+
"loss": 0.0614,
|
| 506 |
+
"step": 690
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.0448,
|
| 510 |
+
"grad_norm": 1.9188823699951172,
|
| 511 |
+
"learning_rate": 7.456000000000001e-06,
|
| 512 |
+
"loss": 0.0539,
|
| 513 |
+
"step": 700
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.04544,
|
| 517 |
+
"grad_norm": 2.0834364891052246,
|
| 518 |
+
"learning_rate": 7.562666666666667e-06,
|
| 519 |
+
"loss": 0.0545,
|
| 520 |
+
"step": 710
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.04608,
|
| 524 |
+
"grad_norm": 2.081609010696411,
|
| 525 |
+
"learning_rate": 7.669333333333335e-06,
|
| 526 |
+
"loss": 0.0563,
|
| 527 |
+
"step": 720
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.04672,
|
| 531 |
+
"grad_norm": 1.461228370666504,
|
| 532 |
+
"learning_rate": 7.776e-06,
|
| 533 |
+
"loss": 0.0495,
|
| 534 |
+
"step": 730
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.04736,
|
| 538 |
+
"grad_norm": 2.7746951580047607,
|
| 539 |
+
"learning_rate": 7.882666666666667e-06,
|
| 540 |
+
"loss": 0.0517,
|
| 541 |
+
"step": 740
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.048,
|
| 545 |
+
"grad_norm": 1.2801299095153809,
|
| 546 |
+
"learning_rate": 7.989333333333335e-06,
|
| 547 |
+
"loss": 0.034,
|
| 548 |
+
"step": 750
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.04864,
|
| 552 |
+
"grad_norm": 3.921491861343384,
|
| 553 |
+
"learning_rate": 8.096e-06,
|
| 554 |
+
"loss": 0.0537,
|
| 555 |
+
"step": 760
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.04928,
|
| 559 |
+
"grad_norm": 1.9986109733581543,
|
| 560 |
+
"learning_rate": 8.202666666666667e-06,
|
| 561 |
+
"loss": 0.0444,
|
| 562 |
+
"step": 770
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.04992,
|
| 566 |
+
"grad_norm": 4.678001403808594,
|
| 567 |
+
"learning_rate": 8.309333333333334e-06,
|
| 568 |
+
"loss": 0.0589,
|
| 569 |
+
"step": 780
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.05056,
|
| 573 |
+
"grad_norm": 1.7302327156066895,
|
| 574 |
+
"learning_rate": 8.416e-06,
|
| 575 |
+
"loss": 0.043,
|
| 576 |
+
"step": 790
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.0512,
|
| 580 |
+
"grad_norm": 2.5643601417541504,
|
| 581 |
+
"learning_rate": 8.522666666666667e-06,
|
| 582 |
+
"loss": 0.0671,
|
| 583 |
+
"step": 800
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.05184,
|
| 587 |
+
"grad_norm": 2.8869082927703857,
|
| 588 |
+
"learning_rate": 8.629333333333334e-06,
|
| 589 |
+
"loss": 0.0565,
|
| 590 |
+
"step": 810
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.05248,
|
| 594 |
+
"grad_norm": 3.723088264465332,
|
| 595 |
+
"learning_rate": 8.736e-06,
|
| 596 |
+
"loss": 0.0428,
|
| 597 |
+
"step": 820
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.05312,
|
| 601 |
+
"grad_norm": 4.387404441833496,
|
| 602 |
+
"learning_rate": 8.842666666666667e-06,
|
| 603 |
+
"loss": 0.0559,
|
| 604 |
+
"step": 830
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.05376,
|
| 608 |
+
"grad_norm": 1.889413833618164,
|
| 609 |
+
"learning_rate": 8.949333333333334e-06,
|
| 610 |
+
"loss": 0.0405,
|
| 611 |
+
"step": 840
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.0544,
|
| 615 |
+
"grad_norm": 1.4773402214050293,
|
| 616 |
+
"learning_rate": 9.056000000000001e-06,
|
| 617 |
+
"loss": 0.0414,
|
| 618 |
+
"step": 850
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.05504,
|
| 622 |
+
"grad_norm": 0.4495583176612854,
|
| 623 |
+
"learning_rate": 9.162666666666668e-06,
|
| 624 |
+
"loss": 0.0446,
|
| 625 |
+
"step": 860
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.05568,
|
| 629 |
+
"grad_norm": 1.158181071281433,
|
| 630 |
+
"learning_rate": 9.269333333333335e-06,
|
| 631 |
+
"loss": 0.0432,
|
| 632 |
+
"step": 870
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.05632,
|
| 636 |
+
"grad_norm": 2.434063196182251,
|
| 637 |
+
"learning_rate": 9.376000000000001e-06,
|
| 638 |
+
"loss": 0.0502,
|
| 639 |
+
"step": 880
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.05696,
|
| 643 |
+
"grad_norm": 1.2009385824203491,
|
| 644 |
+
"learning_rate": 9.482666666666668e-06,
|
| 645 |
+
"loss": 0.0473,
|
| 646 |
+
"step": 890
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.0576,
|
| 650 |
+
"grad_norm": 3.9361023902893066,
|
| 651 |
+
"learning_rate": 9.589333333333333e-06,
|
| 652 |
+
"loss": 0.046,
|
| 653 |
+
"step": 900
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.05824,
|
| 657 |
+
"grad_norm": 2.926988124847412,
|
| 658 |
+
"learning_rate": 9.696000000000002e-06,
|
| 659 |
+
"loss": 0.0464,
|
| 660 |
+
"step": 910
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.05888,
|
| 664 |
+
"grad_norm": 2.215726375579834,
|
| 665 |
+
"learning_rate": 9.802666666666667e-06,
|
| 666 |
+
"loss": 0.0471,
|
| 667 |
+
"step": 920
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.05952,
|
| 671 |
+
"grad_norm": 2.4104645252227783,
|
| 672 |
+
"learning_rate": 9.909333333333334e-06,
|
| 673 |
+
"loss": 0.0482,
|
| 674 |
+
"step": 930
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.06016,
|
| 678 |
+
"grad_norm": 5.664385795593262,
|
| 679 |
+
"learning_rate": 1.0016000000000002e-05,
|
| 680 |
+
"loss": 0.044,
|
| 681 |
+
"step": 940
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.0608,
|
| 685 |
+
"grad_norm": 1.7977851629257202,
|
| 686 |
+
"learning_rate": 1.0122666666666667e-05,
|
| 687 |
+
"loss": 0.0429,
|
| 688 |
+
"step": 950
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.06144,
|
| 692 |
+
"grad_norm": 2.953789472579956,
|
| 693 |
+
"learning_rate": 1.0229333333333334e-05,
|
| 694 |
+
"loss": 0.0533,
|
| 695 |
+
"step": 960
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.06208,
|
| 699 |
+
"grad_norm": 1.6272608041763306,
|
| 700 |
+
"learning_rate": 1.0336000000000002e-05,
|
| 701 |
+
"loss": 0.0307,
|
| 702 |
+
"step": 970
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.06272,
|
| 706 |
+
"grad_norm": 2.8621678352355957,
|
| 707 |
+
"learning_rate": 1.0442666666666667e-05,
|
| 708 |
+
"loss": 0.0461,
|
| 709 |
+
"step": 980
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.06336,
|
| 713 |
+
"grad_norm": 3.70019793510437,
|
| 714 |
+
"learning_rate": 1.0549333333333334e-05,
|
| 715 |
+
"loss": 0.0546,
|
| 716 |
+
"step": 990
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.064,
|
| 720 |
+
"grad_norm": 1.1777608394622803,
|
| 721 |
+
"learning_rate": 1.0656000000000003e-05,
|
| 722 |
+
"loss": 0.0508,
|
| 723 |
+
"step": 1000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.064,
|
| 727 |
+
"eval_accuracy": 0.96139,
|
| 728 |
+
"eval_f1_weighted": 0.9613900534763816,
|
| 729 |
+
"eval_loss": 0.031560152769088745,
|
| 730 |
+
"eval_macro_f1": 0.9613078781032101,
|
| 731 |
+
"eval_precision": 0.9613901086866451,
|
| 732 |
+
"eval_recall": 0.96139,
|
| 733 |
+
"eval_runtime": 1630.3285,
|
| 734 |
+
"eval_samples_per_second": 61.337,
|
| 735 |
+
"eval_steps_per_second": 0.959,
|
| 736 |
+
"step": 1000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.06464,
|
| 740 |
+
"grad_norm": 1.070363163948059,
|
| 741 |
+
"learning_rate": 1.0762666666666668e-05,
|
| 742 |
+
"loss": 0.0383,
|
| 743 |
+
"step": 1010
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.06528,
|
| 747 |
+
"grad_norm": 1.6929458379745483,
|
| 748 |
+
"learning_rate": 1.0869333333333335e-05,
|
| 749 |
+
"loss": 0.0411,
|
| 750 |
+
"step": 1020
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.06592,
|
| 754 |
+
"grad_norm": 3.4057459831237793,
|
| 755 |
+
"learning_rate": 1.0976e-05,
|
| 756 |
+
"loss": 0.0366,
|
| 757 |
+
"step": 1030
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.06656,
|
| 761 |
+
"grad_norm": 4.279348850250244,
|
| 762 |
+
"learning_rate": 1.1082666666666668e-05,
|
| 763 |
+
"loss": 0.0518,
|
| 764 |
+
"step": 1040
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.0672,
|
| 768 |
+
"grad_norm": 2.8239481449127197,
|
| 769 |
+
"learning_rate": 1.1189333333333335e-05,
|
| 770 |
+
"loss": 0.0628,
|
| 771 |
+
"step": 1050
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.06784,
|
| 775 |
+
"grad_norm": 0.6896284222602844,
|
| 776 |
+
"learning_rate": 1.1296e-05,
|
| 777 |
+
"loss": 0.0326,
|
| 778 |
+
"step": 1060
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.06848,
|
| 782 |
+
"grad_norm": 2.390023946762085,
|
| 783 |
+
"learning_rate": 1.1402666666666668e-05,
|
| 784 |
+
"loss": 0.023,
|
| 785 |
+
"step": 1070
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.06912,
|
| 789 |
+
"grad_norm": 4.928287506103516,
|
| 790 |
+
"learning_rate": 1.1509333333333335e-05,
|
| 791 |
+
"loss": 0.054,
|
| 792 |
+
"step": 1080
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.06976,
|
| 796 |
+
"grad_norm": 0.56038498878479,
|
| 797 |
+
"learning_rate": 1.1616e-05,
|
| 798 |
+
"loss": 0.0455,
|
| 799 |
+
"step": 1090
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.0704,
|
| 803 |
+
"grad_norm": 1.9952913522720337,
|
| 804 |
+
"learning_rate": 1.1722666666666667e-05,
|
| 805 |
+
"loss": 0.0498,
|
| 806 |
+
"step": 1100
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.07104,
|
| 810 |
+
"grad_norm": 3.672297239303589,
|
| 811 |
+
"learning_rate": 1.1829333333333335e-05,
|
| 812 |
+
"loss": 0.0316,
|
| 813 |
+
"step": 1110
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.07168,
|
| 817 |
+
"grad_norm": 1.2902151346206665,
|
| 818 |
+
"learning_rate": 1.1936e-05,
|
| 819 |
+
"loss": 0.0307,
|
| 820 |
+
"step": 1120
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.07232,
|
| 824 |
+
"grad_norm": 3.4965603351593018,
|
| 825 |
+
"learning_rate": 1.2042666666666667e-05,
|
| 826 |
+
"loss": 0.0276,
|
| 827 |
+
"step": 1130
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.07296,
|
| 831 |
+
"grad_norm": 2.104722261428833,
|
| 832 |
+
"learning_rate": 1.2149333333333336e-05,
|
| 833 |
+
"loss": 0.0418,
|
| 834 |
+
"step": 1140
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.0736,
|
| 838 |
+
"grad_norm": 3.112037420272827,
|
| 839 |
+
"learning_rate": 1.2256000000000001e-05,
|
| 840 |
+
"loss": 0.0273,
|
| 841 |
+
"step": 1150
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.07424,
|
| 845 |
+
"grad_norm": 3.404317855834961,
|
| 846 |
+
"learning_rate": 1.2362666666666668e-05,
|
| 847 |
+
"loss": 0.05,
|
| 848 |
+
"step": 1160
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.07488,
|
| 852 |
+
"grad_norm": 3.429887056350708,
|
| 853 |
+
"learning_rate": 1.2469333333333333e-05,
|
| 854 |
+
"loss": 0.0373,
|
| 855 |
+
"step": 1170
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.07552,
|
| 859 |
+
"grad_norm": 3.0430426597595215,
|
| 860 |
+
"learning_rate": 1.2576000000000001e-05,
|
| 861 |
+
"loss": 0.0416,
|
| 862 |
+
"step": 1180
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.07616,
|
| 866 |
+
"grad_norm": 0.9473117589950562,
|
| 867 |
+
"learning_rate": 1.2682666666666668e-05,
|
| 868 |
+
"loss": 0.0386,
|
| 869 |
+
"step": 1190
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.0768,
|
| 873 |
+
"grad_norm": 3.2402405738830566,
|
| 874 |
+
"learning_rate": 1.2789333333333333e-05,
|
| 875 |
+
"loss": 0.0483,
|
| 876 |
+
"step": 1200
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.07744,
|
| 880 |
+
"grad_norm": 0.6574758291244507,
|
| 881 |
+
"learning_rate": 1.2896000000000002e-05,
|
| 882 |
+
"loss": 0.0325,
|
| 883 |
+
"step": 1210
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.07808,
|
| 887 |
+
"grad_norm": 1.2989708185195923,
|
| 888 |
+
"learning_rate": 1.3002666666666668e-05,
|
| 889 |
+
"loss": 0.0317,
|
| 890 |
+
"step": 1220
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.07872,
|
| 894 |
+
"grad_norm": 2.3091909885406494,
|
| 895 |
+
"learning_rate": 1.3109333333333333e-05,
|
| 896 |
+
"loss": 0.0333,
|
| 897 |
+
"step": 1230
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.07936,
|
| 901 |
+
"grad_norm": 1.9569207429885864,
|
| 902 |
+
"learning_rate": 1.3216000000000002e-05,
|
| 903 |
+
"loss": 0.0551,
|
| 904 |
+
"step": 1240
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.08,
|
| 908 |
+
"grad_norm": 1.9818898439407349,
|
| 909 |
+
"learning_rate": 1.3322666666666669e-05,
|
| 910 |
+
"loss": 0.0456,
|
| 911 |
+
"step": 1250
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.08064,
|
| 915 |
+
"grad_norm": 1.6688885688781738,
|
| 916 |
+
"learning_rate": 1.3429333333333334e-05,
|
| 917 |
+
"loss": 0.0282,
|
| 918 |
+
"step": 1260
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.08128,
|
| 922 |
+
"grad_norm": 3.399576425552368,
|
| 923 |
+
"learning_rate": 1.3536e-05,
|
| 924 |
+
"loss": 0.0272,
|
| 925 |
+
"step": 1270
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.08192,
|
| 929 |
+
"grad_norm": 1.1946886777877808,
|
| 930 |
+
"learning_rate": 1.3642666666666669e-05,
|
| 931 |
+
"loss": 0.0286,
|
| 932 |
+
"step": 1280
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.08256,
|
| 936 |
+
"grad_norm": 0.6542123556137085,
|
| 937 |
+
"learning_rate": 1.3749333333333334e-05,
|
| 938 |
+
"loss": 0.0402,
|
| 939 |
+
"step": 1290
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.0832,
|
| 943 |
+
"grad_norm": 0.3911895453929901,
|
| 944 |
+
"learning_rate": 1.3856e-05,
|
| 945 |
+
"loss": 0.0233,
|
| 946 |
+
"step": 1300
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.08384,
|
| 950 |
+
"grad_norm": 2.623258352279663,
|
| 951 |
+
"learning_rate": 1.396266666666667e-05,
|
| 952 |
+
"loss": 0.032,
|
| 953 |
+
"step": 1310
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.08448,
|
| 957 |
+
"grad_norm": 0.648289144039154,
|
| 958 |
+
"learning_rate": 1.4069333333333334e-05,
|
| 959 |
+
"loss": 0.0228,
|
| 960 |
+
"step": 1320
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.08512,
|
| 964 |
+
"grad_norm": 3.750211000442505,
|
| 965 |
+
"learning_rate": 1.4176000000000001e-05,
|
| 966 |
+
"loss": 0.0502,
|
| 967 |
+
"step": 1330
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.08576,
|
| 971 |
+
"grad_norm": 2.802426815032959,
|
| 972 |
+
"learning_rate": 1.4282666666666666e-05,
|
| 973 |
+
"loss": 0.0287,
|
| 974 |
+
"step": 1340
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.0864,
|
| 978 |
+
"grad_norm": 1.0664188861846924,
|
| 979 |
+
"learning_rate": 1.4389333333333335e-05,
|
| 980 |
+
"loss": 0.0385,
|
| 981 |
+
"step": 1350
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.08704,
|
| 985 |
+
"grad_norm": 3.218820571899414,
|
| 986 |
+
"learning_rate": 1.4496000000000001e-05,
|
| 987 |
+
"loss": 0.0515,
|
| 988 |
+
"step": 1360
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.08768,
|
| 992 |
+
"grad_norm": 1.9482243061065674,
|
| 993 |
+
"learning_rate": 1.4602666666666667e-05,
|
| 994 |
+
"loss": 0.0258,
|
| 995 |
+
"step": 1370
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.08832,
|
| 999 |
+
"grad_norm": 0.695675253868103,
|
| 1000 |
+
"learning_rate": 1.4709333333333335e-05,
|
| 1001 |
+
"loss": 0.0422,
|
| 1002 |
+
"step": 1380
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.08896,
|
| 1006 |
+
"grad_norm": 0.9671708941459656,
|
| 1007 |
+
"learning_rate": 1.4816000000000002e-05,
|
| 1008 |
+
"loss": 0.0319,
|
| 1009 |
+
"step": 1390
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.0896,
|
| 1013 |
+
"grad_norm": 0.8047491908073425,
|
| 1014 |
+
"learning_rate": 1.4922666666666667e-05,
|
| 1015 |
+
"loss": 0.0423,
|
| 1016 |
+
"step": 1400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.09024,
|
| 1020 |
+
"grad_norm": 1.894417643547058,
|
| 1021 |
+
"learning_rate": 1.5029333333333334e-05,
|
| 1022 |
+
"loss": 0.0327,
|
| 1023 |
+
"step": 1410
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.09088,
|
| 1027 |
+
"grad_norm": 3.023747444152832,
|
| 1028 |
+
"learning_rate": 1.5136000000000002e-05,
|
| 1029 |
+
"loss": 0.0356,
|
| 1030 |
+
"step": 1420
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.09152,
|
| 1034 |
+
"grad_norm": 1.0697004795074463,
|
| 1035 |
+
"learning_rate": 1.5242666666666667e-05,
|
| 1036 |
+
"loss": 0.038,
|
| 1037 |
+
"step": 1430
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.09216,
|
| 1041 |
+
"grad_norm": 1.0740805864334106,
|
| 1042 |
+
"learning_rate": 1.5349333333333332e-05,
|
| 1043 |
+
"loss": 0.037,
|
| 1044 |
+
"step": 1440
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.0928,
|
| 1048 |
+
"grad_norm": 1.7463656663894653,
|
| 1049 |
+
"learning_rate": 1.5456000000000002e-05,
|
| 1050 |
+
"loss": 0.0277,
|
| 1051 |
+
"step": 1450
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.09344,
|
| 1055 |
+
"grad_norm": 0.6604284644126892,
|
| 1056 |
+
"learning_rate": 1.556266666666667e-05,
|
| 1057 |
+
"loss": 0.0259,
|
| 1058 |
+
"step": 1460
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.09408,
|
| 1062 |
+
"grad_norm": 2.2307727336883545,
|
| 1063 |
+
"learning_rate": 1.5669333333333333e-05,
|
| 1064 |
+
"loss": 0.0311,
|
| 1065 |
+
"step": 1470
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.09472,
|
| 1069 |
+
"grad_norm": 0.47977620363235474,
|
| 1070 |
+
"learning_rate": 1.5776e-05,
|
| 1071 |
+
"loss": 0.0184,
|
| 1072 |
+
"step": 1480
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.09536,
|
| 1076 |
+
"grad_norm": 1.1453766822814941,
|
| 1077 |
+
"learning_rate": 1.588266666666667e-05,
|
| 1078 |
+
"loss": 0.0303,
|
| 1079 |
+
"step": 1490
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.096,
|
| 1083 |
+
"grad_norm": 2.8183929920196533,
|
| 1084 |
+
"learning_rate": 1.5989333333333333e-05,
|
| 1085 |
+
"loss": 0.0498,
|
| 1086 |
+
"step": 1500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.096,
|
| 1090 |
+
"eval_accuracy": 0.96992,
|
| 1091 |
+
"eval_f1_weighted": 0.9699323760484765,
|
| 1092 |
+
"eval_loss": 0.029355797916650772,
|
| 1093 |
+
"eval_macro_f1": 0.9698851462608304,
|
| 1094 |
+
"eval_precision": 0.9702279455514112,
|
| 1095 |
+
"eval_recall": 0.96992,
|
| 1096 |
+
"eval_runtime": 1629.7849,
|
| 1097 |
+
"eval_samples_per_second": 61.358,
|
| 1098 |
+
"eval_steps_per_second": 0.959,
|
| 1099 |
+
"step": 1500
|
| 1100 |
+
}
|
| 1101 |
+
],
|
| 1102 |
+
"logging_steps": 10,
|
| 1103 |
+
"max_steps": 15625,
|
| 1104 |
+
"num_input_tokens_seen": 0,
|
| 1105 |
+
"num_train_epochs": 1,
|
| 1106 |
+
"save_steps": 500,
|
| 1107 |
+
"stateful_callbacks": {
|
| 1108 |
+
"EarlyStoppingCallback": {
|
| 1109 |
+
"args": {
|
| 1110 |
+
"early_stopping_patience": 3,
|
| 1111 |
+
"early_stopping_threshold": 0.0
|
| 1112 |
+
},
|
| 1113 |
+
"attributes": {
|
| 1114 |
+
"early_stopping_patience_counter": 0
|
| 1115 |
+
}
|
| 1116 |
+
},
|
| 1117 |
+
"TrainerControl": {
|
| 1118 |
+
"args": {
|
| 1119 |
+
"should_epoch_stop": false,
|
| 1120 |
+
"should_evaluate": false,
|
| 1121 |
+
"should_log": false,
|
| 1122 |
+
"should_save": true,
|
| 1123 |
+
"should_training_stop": false
|
| 1124 |
+
},
|
| 1125 |
+
"attributes": {}
|
| 1126 |
+
}
|
| 1127 |
+
},
|
| 1128 |
+
"total_flos": 1.262245688095872e+16,
|
| 1129 |
+
"train_batch_size": 32,
|
| 1130 |
+
"trial_name": null,
|
| 1131 |
+
"trial_params": null
|
| 1132 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0256e15114b0df6f165bb3e6d75f279805d6f125574ac927fca7a602f9449f4a
|
| 3 |
+
size 5777
|
checkpoint-1500/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|