Upload folder using huggingface_hub
Browse files- checkpoints/checkpoint-3800/config.json +26 -0
- checkpoints/checkpoint-3800/model.safetensors +3 -0
- checkpoints/checkpoint-3800/optimizer.pt +3 -0
- checkpoints/checkpoint-3800/rng_state.pth +3 -0
- checkpoints/checkpoint-3800/scheduler.pt +3 -0
- checkpoints/checkpoint-3800/special_tokens_map.json +7 -0
- checkpoints/checkpoint-3800/tokenizer.json +0 -0
- checkpoints/checkpoint-3800/tokenizer_config.json +56 -0
- checkpoints/checkpoint-3800/trainer_state.json +425 -0
- checkpoints/checkpoint-3800/training_args.bin +3 -0
- checkpoints/checkpoint-3800/vocab.txt +0 -0
- checkpoints/checkpoint-9600/config.json +26 -0
- checkpoints/checkpoint-9600/model.safetensors +3 -0
- checkpoints/checkpoint-9600/optimizer.pt +3 -0
- checkpoints/checkpoint-9600/rng_state.pth +3 -0
- checkpoints/checkpoint-9600/scheduler.pt +3 -0
- checkpoints/checkpoint-9600/special_tokens_map.json +7 -0
- checkpoints/checkpoint-9600/tokenizer.json +0 -0
- checkpoints/checkpoint-9600/tokenizer_config.json +56 -0
- checkpoints/checkpoint-9600/trainer_state.json +1031 -0
- checkpoints/checkpoint-9600/training_args.bin +3 -0
- checkpoints/checkpoint-9600/vocab.txt +0 -0
- final_model.pt +3 -0
- final_model_meta.json +1 -0
- test_outputs.jsonl +96 -0
checkpoints/checkpoint-3800/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
checkpoints/checkpoint-3800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb024fd00e2f293ad9457ef7a597b0668ff451a3cb49af4307df50cdfcb0bc9d
|
| 3 |
+
size 437958648
|
checkpoints/checkpoint-3800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e21dd4905d7d4b4b7a2f11afd2cbbd30812a8bcc01901d870ba7ff2665fd69de
|
| 3 |
+
size 876038394
|
checkpoints/checkpoint-3800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36c48f8e80e8a5ae133fa1a36f4a852e5d2cf6601b2de8db40115f86eaa42ff7
|
| 3 |
+
size 14244
|
checkpoints/checkpoint-3800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a13e2ed1331335e98f400adbf34a4d2528fb3cba2a4951ab0d96b6012eac4bc9
|
| 3 |
+
size 1064
|
checkpoints/checkpoint-3800/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoints/checkpoint-3800/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-3800/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoints/checkpoint-3800/trainer_state.json
ADDED
|
@@ -0,0 +1,425 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3800,
|
| 3 |
+
"best_metric": 0.940625,
|
| 4 |
+
"best_model_checkpoint": "../merge_models/yelp_amzn/task2/checkpoints/checkpoint-3800",
|
| 5 |
+
"epoch": 1.5833333333333335,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 3800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.041666666666666664,
|
| 14 |
+
"eval_accuracy": 0.89375,
|
| 15 |
+
"eval_loss": 0.4754180908203125,
|
| 16 |
+
"eval_runtime": 2.8838,
|
| 17 |
+
"eval_samples_per_second": 110.967,
|
| 18 |
+
"eval_steps_per_second": 55.483,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.08333333333333333,
|
| 23 |
+
"eval_accuracy": 0.84375,
|
| 24 |
+
"eval_loss": 0.7783083319664001,
|
| 25 |
+
"eval_runtime": 2.8722,
|
| 26 |
+
"eval_samples_per_second": 111.413,
|
| 27 |
+
"eval_steps_per_second": 55.706,
|
| 28 |
+
"step": 200
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.125,
|
| 32 |
+
"eval_accuracy": 0.91875,
|
| 33 |
+
"eval_loss": 0.4844598174095154,
|
| 34 |
+
"eval_runtime": 2.9679,
|
| 35 |
+
"eval_samples_per_second": 107.821,
|
| 36 |
+
"eval_steps_per_second": 53.911,
|
| 37 |
+
"step": 300
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"eval_accuracy": 0.8875,
|
| 42 |
+
"eval_loss": 0.5653066635131836,
|
| 43 |
+
"eval_runtime": 2.8114,
|
| 44 |
+
"eval_samples_per_second": 113.822,
|
| 45 |
+
"eval_steps_per_second": 56.911,
|
| 46 |
+
"step": 400
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.20833333333333334,
|
| 50 |
+
"grad_norm": 0.08967381715774536,
|
| 51 |
+
"learning_rate": 3.792083333333334e-05,
|
| 52 |
+
"loss": 0.5352,
|
| 53 |
+
"step": 500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.20833333333333334,
|
| 57 |
+
"eval_accuracy": 0.89375,
|
| 58 |
+
"eval_loss": 0.4834604263305664,
|
| 59 |
+
"eval_runtime": 3.0663,
|
| 60 |
+
"eval_samples_per_second": 104.361,
|
| 61 |
+
"eval_steps_per_second": 52.18,
|
| 62 |
+
"step": 500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"eval_accuracy": 0.928125,
|
| 67 |
+
"eval_loss": 0.3776414394378662,
|
| 68 |
+
"eval_runtime": 2.9683,
|
| 69 |
+
"eval_samples_per_second": 107.806,
|
| 70 |
+
"eval_steps_per_second": 53.903,
|
| 71 |
+
"step": 600
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.2916666666666667,
|
| 75 |
+
"eval_accuracy": 0.846875,
|
| 76 |
+
"eval_loss": 0.7465925216674805,
|
| 77 |
+
"eval_runtime": 3.1091,
|
| 78 |
+
"eval_samples_per_second": 102.924,
|
| 79 |
+
"eval_steps_per_second": 51.462,
|
| 80 |
+
"step": 700
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.3333333333333333,
|
| 84 |
+
"eval_accuracy": 0.91875,
|
| 85 |
+
"eval_loss": 0.43123188614845276,
|
| 86 |
+
"eval_runtime": 2.9983,
|
| 87 |
+
"eval_samples_per_second": 106.727,
|
| 88 |
+
"eval_steps_per_second": 53.363,
|
| 89 |
+
"step": 800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.375,
|
| 93 |
+
"eval_accuracy": 0.921875,
|
| 94 |
+
"eval_loss": 0.3499329686164856,
|
| 95 |
+
"eval_runtime": 3.1929,
|
| 96 |
+
"eval_samples_per_second": 100.224,
|
| 97 |
+
"eval_steps_per_second": 50.112,
|
| 98 |
+
"step": 900
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.4166666666666667,
|
| 102 |
+
"grad_norm": 30.974872589111328,
|
| 103 |
+
"learning_rate": 3.5837500000000004e-05,
|
| 104 |
+
"loss": 0.5275,
|
| 105 |
+
"step": 1000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.4166666666666667,
|
| 109 |
+
"eval_accuracy": 0.834375,
|
| 110 |
+
"eval_loss": 0.7541207671165466,
|
| 111 |
+
"eval_runtime": 2.9939,
|
| 112 |
+
"eval_samples_per_second": 106.885,
|
| 113 |
+
"eval_steps_per_second": 53.442,
|
| 114 |
+
"step": 1000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4583333333333333,
|
| 118 |
+
"eval_accuracy": 0.909375,
|
| 119 |
+
"eval_loss": 0.47846031188964844,
|
| 120 |
+
"eval_runtime": 2.965,
|
| 121 |
+
"eval_samples_per_second": 107.926,
|
| 122 |
+
"eval_steps_per_second": 53.963,
|
| 123 |
+
"step": 1100
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.5,
|
| 127 |
+
"eval_accuracy": 0.903125,
|
| 128 |
+
"eval_loss": 0.6214954257011414,
|
| 129 |
+
"eval_runtime": 2.9259,
|
| 130 |
+
"eval_samples_per_second": 109.368,
|
| 131 |
+
"eval_steps_per_second": 54.684,
|
| 132 |
+
"step": 1200
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.5416666666666666,
|
| 136 |
+
"eval_accuracy": 0.915625,
|
| 137 |
+
"eval_loss": 0.42658787965774536,
|
| 138 |
+
"eval_runtime": 3.0223,
|
| 139 |
+
"eval_samples_per_second": 105.879,
|
| 140 |
+
"eval_steps_per_second": 52.939,
|
| 141 |
+
"step": 1300
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.5833333333333334,
|
| 145 |
+
"eval_accuracy": 0.90625,
|
| 146 |
+
"eval_loss": 0.3985453248023987,
|
| 147 |
+
"eval_runtime": 2.9117,
|
| 148 |
+
"eval_samples_per_second": 109.903,
|
| 149 |
+
"eval_steps_per_second": 54.952,
|
| 150 |
+
"step": 1400
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.625,
|
| 154 |
+
"grad_norm": 0.9175795316696167,
|
| 155 |
+
"learning_rate": 3.375416666666667e-05,
|
| 156 |
+
"loss": 0.498,
|
| 157 |
+
"step": 1500
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.625,
|
| 161 |
+
"eval_accuracy": 0.9,
|
| 162 |
+
"eval_loss": 0.5395756363868713,
|
| 163 |
+
"eval_runtime": 2.8898,
|
| 164 |
+
"eval_samples_per_second": 110.733,
|
| 165 |
+
"eval_steps_per_second": 55.366,
|
| 166 |
+
"step": 1500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.6666666666666666,
|
| 170 |
+
"eval_accuracy": 0.90625,
|
| 171 |
+
"eval_loss": 0.48287898302078247,
|
| 172 |
+
"eval_runtime": 3.07,
|
| 173 |
+
"eval_samples_per_second": 104.235,
|
| 174 |
+
"eval_steps_per_second": 52.118,
|
| 175 |
+
"step": 1600
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.7083333333333334,
|
| 179 |
+
"eval_accuracy": 0.915625,
|
| 180 |
+
"eval_loss": 0.4366416037082672,
|
| 181 |
+
"eval_runtime": 3.0743,
|
| 182 |
+
"eval_samples_per_second": 104.09,
|
| 183 |
+
"eval_steps_per_second": 52.045,
|
| 184 |
+
"step": 1700
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.75,
|
| 188 |
+
"eval_accuracy": 0.9125,
|
| 189 |
+
"eval_loss": 0.5095136761665344,
|
| 190 |
+
"eval_runtime": 3.0831,
|
| 191 |
+
"eval_samples_per_second": 103.792,
|
| 192 |
+
"eval_steps_per_second": 51.896,
|
| 193 |
+
"step": 1800
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.7916666666666666,
|
| 197 |
+
"eval_accuracy": 0.915625,
|
| 198 |
+
"eval_loss": 0.4809819161891937,
|
| 199 |
+
"eval_runtime": 2.8721,
|
| 200 |
+
"eval_samples_per_second": 111.416,
|
| 201 |
+
"eval_steps_per_second": 55.708,
|
| 202 |
+
"step": 1900
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.8333333333333334,
|
| 206 |
+
"grad_norm": 20.087257385253906,
|
| 207 |
+
"learning_rate": 3.1670833333333335e-05,
|
| 208 |
+
"loss": 0.3666,
|
| 209 |
+
"step": 2000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.8333333333333334,
|
| 213 |
+
"eval_accuracy": 0.8875,
|
| 214 |
+
"eval_loss": 0.6681545376777649,
|
| 215 |
+
"eval_runtime": 2.8368,
|
| 216 |
+
"eval_samples_per_second": 112.803,
|
| 217 |
+
"eval_steps_per_second": 56.402,
|
| 218 |
+
"step": 2000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.875,
|
| 222 |
+
"eval_accuracy": 0.9125,
|
| 223 |
+
"eval_loss": 0.4883766770362854,
|
| 224 |
+
"eval_runtime": 3.2206,
|
| 225 |
+
"eval_samples_per_second": 99.36,
|
| 226 |
+
"eval_steps_per_second": 49.68,
|
| 227 |
+
"step": 2100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.9166666666666666,
|
| 231 |
+
"eval_accuracy": 0.915625,
|
| 232 |
+
"eval_loss": 0.37708765268325806,
|
| 233 |
+
"eval_runtime": 3.2053,
|
| 234 |
+
"eval_samples_per_second": 99.835,
|
| 235 |
+
"eval_steps_per_second": 49.917,
|
| 236 |
+
"step": 2200
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.9583333333333334,
|
| 240 |
+
"eval_accuracy": 0.91875,
|
| 241 |
+
"eval_loss": 0.38682788610458374,
|
| 242 |
+
"eval_runtime": 2.8559,
|
| 243 |
+
"eval_samples_per_second": 112.048,
|
| 244 |
+
"eval_steps_per_second": 56.024,
|
| 245 |
+
"step": 2300
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.0,
|
| 249 |
+
"eval_accuracy": 0.90625,
|
| 250 |
+
"eval_loss": 0.5085960626602173,
|
| 251 |
+
"eval_runtime": 2.9762,
|
| 252 |
+
"eval_samples_per_second": 107.518,
|
| 253 |
+
"eval_steps_per_second": 53.759,
|
| 254 |
+
"step": 2400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.0416666666666667,
|
| 258 |
+
"grad_norm": 0.18000628054141998,
|
| 259 |
+
"learning_rate": 2.9587500000000004e-05,
|
| 260 |
+
"loss": 0.3676,
|
| 261 |
+
"step": 2500
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.0416666666666667,
|
| 265 |
+
"eval_accuracy": 0.921875,
|
| 266 |
+
"eval_loss": 0.467485249042511,
|
| 267 |
+
"eval_runtime": 2.8829,
|
| 268 |
+
"eval_samples_per_second": 111.001,
|
| 269 |
+
"eval_steps_per_second": 55.5,
|
| 270 |
+
"step": 2500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.0833333333333333,
|
| 274 |
+
"eval_accuracy": 0.89375,
|
| 275 |
+
"eval_loss": 0.6294034719467163,
|
| 276 |
+
"eval_runtime": 3.211,
|
| 277 |
+
"eval_samples_per_second": 99.656,
|
| 278 |
+
"eval_steps_per_second": 49.828,
|
| 279 |
+
"step": 2600
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.125,
|
| 283 |
+
"eval_accuracy": 0.915625,
|
| 284 |
+
"eval_loss": 0.5321327447891235,
|
| 285 |
+
"eval_runtime": 3.0709,
|
| 286 |
+
"eval_samples_per_second": 104.205,
|
| 287 |
+
"eval_steps_per_second": 52.102,
|
| 288 |
+
"step": 2700
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.1666666666666667,
|
| 292 |
+
"eval_accuracy": 0.903125,
|
| 293 |
+
"eval_loss": 0.5952805280685425,
|
| 294 |
+
"eval_runtime": 2.8172,
|
| 295 |
+
"eval_samples_per_second": 113.588,
|
| 296 |
+
"eval_steps_per_second": 56.794,
|
| 297 |
+
"step": 2800
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.2083333333333333,
|
| 301 |
+
"eval_accuracy": 0.915625,
|
| 302 |
+
"eval_loss": 0.5122720003128052,
|
| 303 |
+
"eval_runtime": 2.9015,
|
| 304 |
+
"eval_samples_per_second": 110.288,
|
| 305 |
+
"eval_steps_per_second": 55.144,
|
| 306 |
+
"step": 2900
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.25,
|
| 310 |
+
"grad_norm": 0.010164022445678711,
|
| 311 |
+
"learning_rate": 2.750416666666667e-05,
|
| 312 |
+
"loss": 0.2205,
|
| 313 |
+
"step": 3000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.25,
|
| 317 |
+
"eval_accuracy": 0.884375,
|
| 318 |
+
"eval_loss": 0.7141415476799011,
|
| 319 |
+
"eval_runtime": 2.8765,
|
| 320 |
+
"eval_samples_per_second": 111.245,
|
| 321 |
+
"eval_steps_per_second": 55.622,
|
| 322 |
+
"step": 3000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.2916666666666667,
|
| 326 |
+
"eval_accuracy": 0.9,
|
| 327 |
+
"eval_loss": 0.5358812212944031,
|
| 328 |
+
"eval_runtime": 3.1535,
|
| 329 |
+
"eval_samples_per_second": 101.475,
|
| 330 |
+
"eval_steps_per_second": 50.737,
|
| 331 |
+
"step": 3100
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.3333333333333333,
|
| 335 |
+
"eval_accuracy": 0.915625,
|
| 336 |
+
"eval_loss": 0.5337697863578796,
|
| 337 |
+
"eval_runtime": 3.1817,
|
| 338 |
+
"eval_samples_per_second": 100.575,
|
| 339 |
+
"eval_steps_per_second": 50.287,
|
| 340 |
+
"step": 3200
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.375,
|
| 344 |
+
"eval_accuracy": 0.915625,
|
| 345 |
+
"eval_loss": 0.5300858020782471,
|
| 346 |
+
"eval_runtime": 2.8486,
|
| 347 |
+
"eval_samples_per_second": 112.336,
|
| 348 |
+
"eval_steps_per_second": 56.168,
|
| 349 |
+
"step": 3300
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.4166666666666667,
|
| 353 |
+
"eval_accuracy": 0.909375,
|
| 354 |
+
"eval_loss": 0.5486552119255066,
|
| 355 |
+
"eval_runtime": 2.9044,
|
| 356 |
+
"eval_samples_per_second": 110.177,
|
| 357 |
+
"eval_steps_per_second": 55.089,
|
| 358 |
+
"step": 3400
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.4583333333333333,
|
| 362 |
+
"grad_norm": 0.026199543848633766,
|
| 363 |
+
"learning_rate": 2.5420833333333336e-05,
|
| 364 |
+
"loss": 0.2262,
|
| 365 |
+
"step": 3500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.4583333333333333,
|
| 369 |
+
"eval_accuracy": 0.921875,
|
| 370 |
+
"eval_loss": 0.4780457615852356,
|
| 371 |
+
"eval_runtime": 3.1987,
|
| 372 |
+
"eval_samples_per_second": 100.041,
|
| 373 |
+
"eval_steps_per_second": 50.02,
|
| 374 |
+
"step": 3500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.5,
|
| 378 |
+
"eval_accuracy": 0.934375,
|
| 379 |
+
"eval_loss": 0.44137653708457947,
|
| 380 |
+
"eval_runtime": 3.1132,
|
| 381 |
+
"eval_samples_per_second": 102.788,
|
| 382 |
+
"eval_steps_per_second": 51.394,
|
| 383 |
+
"step": 3600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.5416666666666665,
|
| 387 |
+
"eval_accuracy": 0.91875,
|
| 388 |
+
"eval_loss": 0.5386074781417847,
|
| 389 |
+
"eval_runtime": 3.0032,
|
| 390 |
+
"eval_samples_per_second": 106.553,
|
| 391 |
+
"eval_steps_per_second": 53.276,
|
| 392 |
+
"step": 3700
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.5833333333333335,
|
| 396 |
+
"eval_accuracy": 0.940625,
|
| 397 |
+
"eval_loss": 0.3621799945831299,
|
| 398 |
+
"eval_runtime": 3.2102,
|
| 399 |
+
"eval_samples_per_second": 99.683,
|
| 400 |
+
"eval_steps_per_second": 49.842,
|
| 401 |
+
"step": 3800
|
| 402 |
+
}
|
| 403 |
+
],
|
| 404 |
+
"logging_steps": 500,
|
| 405 |
+
"max_steps": 9600,
|
| 406 |
+
"num_input_tokens_seen": 0,
|
| 407 |
+
"num_train_epochs": 4,
|
| 408 |
+
"save_steps": 100,
|
| 409 |
+
"stateful_callbacks": {
|
| 410 |
+
"TrainerControl": {
|
| 411 |
+
"args": {
|
| 412 |
+
"should_epoch_stop": false,
|
| 413 |
+
"should_evaluate": false,
|
| 414 |
+
"should_log": false,
|
| 415 |
+
"should_save": true,
|
| 416 |
+
"should_training_stop": false
|
| 417 |
+
},
|
| 418 |
+
"attributes": {}
|
| 419 |
+
}
|
| 420 |
+
},
|
| 421 |
+
"total_flos": 529497637803720.0,
|
| 422 |
+
"train_batch_size": 2,
|
| 423 |
+
"trial_name": null,
|
| 424 |
+
"trial_params": null
|
| 425 |
+
}
|
checkpoints/checkpoint-3800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e993d463aba6969e9f6ddada547c66913d99677754065e5d3d85907f9219bf4
|
| 3 |
+
size 5304
|
checkpoints/checkpoint-3800/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-9600/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
checkpoints/checkpoint-9600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1538789105abd9e475d3a38984ca1e5f1d795bb6c0eaf4570e7961ad8d0a7f07
|
| 3 |
+
size 437958648
|
checkpoints/checkpoint-9600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9788efc00046fc8fed7efc876bd061784edd26cc384ab4358588044d76cd923d
|
| 3 |
+
size 876038394
|
checkpoints/checkpoint-9600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09950b0ca2e90173d680f1fdb9a5a7a538916c1b6ee6698cc2a8d18382b7b5b9
|
| 3 |
+
size 14244
|
checkpoints/checkpoint-9600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70b4b49382b81ae82c82f376964b1763ea6f54b6fa49526da8f199bd9ab93c7a
|
| 3 |
+
size 1064
|
checkpoints/checkpoint-9600/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoints/checkpoint-9600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints/checkpoint-9600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "BertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoints/checkpoint-9600/trainer_state.json
ADDED
|
@@ -0,0 +1,1031 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3800,
|
| 3 |
+
"best_metric": 0.940625,
|
| 4 |
+
"best_model_checkpoint": "../merge_models/yelp_amzn/task2/checkpoints/checkpoint-3800",
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 9600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.041666666666666664,
|
| 14 |
+
"eval_accuracy": 0.89375,
|
| 15 |
+
"eval_loss": 0.4754180908203125,
|
| 16 |
+
"eval_runtime": 2.8838,
|
| 17 |
+
"eval_samples_per_second": 110.967,
|
| 18 |
+
"eval_steps_per_second": 55.483,
|
| 19 |
+
"step": 100
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.08333333333333333,
|
| 23 |
+
"eval_accuracy": 0.84375,
|
| 24 |
+
"eval_loss": 0.7783083319664001,
|
| 25 |
+
"eval_runtime": 2.8722,
|
| 26 |
+
"eval_samples_per_second": 111.413,
|
| 27 |
+
"eval_steps_per_second": 55.706,
|
| 28 |
+
"step": 200
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.125,
|
| 32 |
+
"eval_accuracy": 0.91875,
|
| 33 |
+
"eval_loss": 0.4844598174095154,
|
| 34 |
+
"eval_runtime": 2.9679,
|
| 35 |
+
"eval_samples_per_second": 107.821,
|
| 36 |
+
"eval_steps_per_second": 53.911,
|
| 37 |
+
"step": 300
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"eval_accuracy": 0.8875,
|
| 42 |
+
"eval_loss": 0.5653066635131836,
|
| 43 |
+
"eval_runtime": 2.8114,
|
| 44 |
+
"eval_samples_per_second": 113.822,
|
| 45 |
+
"eval_steps_per_second": 56.911,
|
| 46 |
+
"step": 400
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.20833333333333334,
|
| 50 |
+
"grad_norm": 0.08967381715774536,
|
| 51 |
+
"learning_rate": 3.792083333333334e-05,
|
| 52 |
+
"loss": 0.5352,
|
| 53 |
+
"step": 500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.20833333333333334,
|
| 57 |
+
"eval_accuracy": 0.89375,
|
| 58 |
+
"eval_loss": 0.4834604263305664,
|
| 59 |
+
"eval_runtime": 3.0663,
|
| 60 |
+
"eval_samples_per_second": 104.361,
|
| 61 |
+
"eval_steps_per_second": 52.18,
|
| 62 |
+
"step": 500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"eval_accuracy": 0.928125,
|
| 67 |
+
"eval_loss": 0.3776414394378662,
|
| 68 |
+
"eval_runtime": 2.9683,
|
| 69 |
+
"eval_samples_per_second": 107.806,
|
| 70 |
+
"eval_steps_per_second": 53.903,
|
| 71 |
+
"step": 600
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.2916666666666667,
|
| 75 |
+
"eval_accuracy": 0.846875,
|
| 76 |
+
"eval_loss": 0.7465925216674805,
|
| 77 |
+
"eval_runtime": 3.1091,
|
| 78 |
+
"eval_samples_per_second": 102.924,
|
| 79 |
+
"eval_steps_per_second": 51.462,
|
| 80 |
+
"step": 700
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.3333333333333333,
|
| 84 |
+
"eval_accuracy": 0.91875,
|
| 85 |
+
"eval_loss": 0.43123188614845276,
|
| 86 |
+
"eval_runtime": 2.9983,
|
| 87 |
+
"eval_samples_per_second": 106.727,
|
| 88 |
+
"eval_steps_per_second": 53.363,
|
| 89 |
+
"step": 800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.375,
|
| 93 |
+
"eval_accuracy": 0.921875,
|
| 94 |
+
"eval_loss": 0.3499329686164856,
|
| 95 |
+
"eval_runtime": 3.1929,
|
| 96 |
+
"eval_samples_per_second": 100.224,
|
| 97 |
+
"eval_steps_per_second": 50.112,
|
| 98 |
+
"step": 900
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.4166666666666667,
|
| 102 |
+
"grad_norm": 30.974872589111328,
|
| 103 |
+
"learning_rate": 3.5837500000000004e-05,
|
| 104 |
+
"loss": 0.5275,
|
| 105 |
+
"step": 1000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.4166666666666667,
|
| 109 |
+
"eval_accuracy": 0.834375,
|
| 110 |
+
"eval_loss": 0.7541207671165466,
|
| 111 |
+
"eval_runtime": 2.9939,
|
| 112 |
+
"eval_samples_per_second": 106.885,
|
| 113 |
+
"eval_steps_per_second": 53.442,
|
| 114 |
+
"step": 1000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4583333333333333,
|
| 118 |
+
"eval_accuracy": 0.909375,
|
| 119 |
+
"eval_loss": 0.47846031188964844,
|
| 120 |
+
"eval_runtime": 2.965,
|
| 121 |
+
"eval_samples_per_second": 107.926,
|
| 122 |
+
"eval_steps_per_second": 53.963,
|
| 123 |
+
"step": 1100
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.5,
|
| 127 |
+
"eval_accuracy": 0.903125,
|
| 128 |
+
"eval_loss": 0.6214954257011414,
|
| 129 |
+
"eval_runtime": 2.9259,
|
| 130 |
+
"eval_samples_per_second": 109.368,
|
| 131 |
+
"eval_steps_per_second": 54.684,
|
| 132 |
+
"step": 1200
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.5416666666666666,
|
| 136 |
+
"eval_accuracy": 0.915625,
|
| 137 |
+
"eval_loss": 0.42658787965774536,
|
| 138 |
+
"eval_runtime": 3.0223,
|
| 139 |
+
"eval_samples_per_second": 105.879,
|
| 140 |
+
"eval_steps_per_second": 52.939,
|
| 141 |
+
"step": 1300
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.5833333333333334,
|
| 145 |
+
"eval_accuracy": 0.90625,
|
| 146 |
+
"eval_loss": 0.3985453248023987,
|
| 147 |
+
"eval_runtime": 2.9117,
|
| 148 |
+
"eval_samples_per_second": 109.903,
|
| 149 |
+
"eval_steps_per_second": 54.952,
|
| 150 |
+
"step": 1400
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.625,
|
| 154 |
+
"grad_norm": 0.9175795316696167,
|
| 155 |
+
"learning_rate": 3.375416666666667e-05,
|
| 156 |
+
"loss": 0.498,
|
| 157 |
+
"step": 1500
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.625,
|
| 161 |
+
"eval_accuracy": 0.9,
|
| 162 |
+
"eval_loss": 0.5395756363868713,
|
| 163 |
+
"eval_runtime": 2.8898,
|
| 164 |
+
"eval_samples_per_second": 110.733,
|
| 165 |
+
"eval_steps_per_second": 55.366,
|
| 166 |
+
"step": 1500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.6666666666666666,
|
| 170 |
+
"eval_accuracy": 0.90625,
|
| 171 |
+
"eval_loss": 0.48287898302078247,
|
| 172 |
+
"eval_runtime": 3.07,
|
| 173 |
+
"eval_samples_per_second": 104.235,
|
| 174 |
+
"eval_steps_per_second": 52.118,
|
| 175 |
+
"step": 1600
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.7083333333333334,
|
| 179 |
+
"eval_accuracy": 0.915625,
|
| 180 |
+
"eval_loss": 0.4366416037082672,
|
| 181 |
+
"eval_runtime": 3.0743,
|
| 182 |
+
"eval_samples_per_second": 104.09,
|
| 183 |
+
"eval_steps_per_second": 52.045,
|
| 184 |
+
"step": 1700
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.75,
|
| 188 |
+
"eval_accuracy": 0.9125,
|
| 189 |
+
"eval_loss": 0.5095136761665344,
|
| 190 |
+
"eval_runtime": 3.0831,
|
| 191 |
+
"eval_samples_per_second": 103.792,
|
| 192 |
+
"eval_steps_per_second": 51.896,
|
| 193 |
+
"step": 1800
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.7916666666666666,
|
| 197 |
+
"eval_accuracy": 0.915625,
|
| 198 |
+
"eval_loss": 0.4809819161891937,
|
| 199 |
+
"eval_runtime": 2.8721,
|
| 200 |
+
"eval_samples_per_second": 111.416,
|
| 201 |
+
"eval_steps_per_second": 55.708,
|
| 202 |
+
"step": 1900
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.8333333333333334,
|
| 206 |
+
"grad_norm": 20.087257385253906,
|
| 207 |
+
"learning_rate": 3.1670833333333335e-05,
|
| 208 |
+
"loss": 0.3666,
|
| 209 |
+
"step": 2000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.8333333333333334,
|
| 213 |
+
"eval_accuracy": 0.8875,
|
| 214 |
+
"eval_loss": 0.6681545376777649,
|
| 215 |
+
"eval_runtime": 2.8368,
|
| 216 |
+
"eval_samples_per_second": 112.803,
|
| 217 |
+
"eval_steps_per_second": 56.402,
|
| 218 |
+
"step": 2000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.875,
|
| 222 |
+
"eval_accuracy": 0.9125,
|
| 223 |
+
"eval_loss": 0.4883766770362854,
|
| 224 |
+
"eval_runtime": 3.2206,
|
| 225 |
+
"eval_samples_per_second": 99.36,
|
| 226 |
+
"eval_steps_per_second": 49.68,
|
| 227 |
+
"step": 2100
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.9166666666666666,
|
| 231 |
+
"eval_accuracy": 0.915625,
|
| 232 |
+
"eval_loss": 0.37708765268325806,
|
| 233 |
+
"eval_runtime": 3.2053,
|
| 234 |
+
"eval_samples_per_second": 99.835,
|
| 235 |
+
"eval_steps_per_second": 49.917,
|
| 236 |
+
"step": 2200
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.9583333333333334,
|
| 240 |
+
"eval_accuracy": 0.91875,
|
| 241 |
+
"eval_loss": 0.38682788610458374,
|
| 242 |
+
"eval_runtime": 2.8559,
|
| 243 |
+
"eval_samples_per_second": 112.048,
|
| 244 |
+
"eval_steps_per_second": 56.024,
|
| 245 |
+
"step": 2300
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.0,
|
| 249 |
+
"eval_accuracy": 0.90625,
|
| 250 |
+
"eval_loss": 0.5085960626602173,
|
| 251 |
+
"eval_runtime": 2.9762,
|
| 252 |
+
"eval_samples_per_second": 107.518,
|
| 253 |
+
"eval_steps_per_second": 53.759,
|
| 254 |
+
"step": 2400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.0416666666666667,
|
| 258 |
+
"grad_norm": 0.18000628054141998,
|
| 259 |
+
"learning_rate": 2.9587500000000004e-05,
|
| 260 |
+
"loss": 0.3676,
|
| 261 |
+
"step": 2500
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.0416666666666667,
|
| 265 |
+
"eval_accuracy": 0.921875,
|
| 266 |
+
"eval_loss": 0.467485249042511,
|
| 267 |
+
"eval_runtime": 2.8829,
|
| 268 |
+
"eval_samples_per_second": 111.001,
|
| 269 |
+
"eval_steps_per_second": 55.5,
|
| 270 |
+
"step": 2500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.0833333333333333,
|
| 274 |
+
"eval_accuracy": 0.89375,
|
| 275 |
+
"eval_loss": 0.6294034719467163,
|
| 276 |
+
"eval_runtime": 3.211,
|
| 277 |
+
"eval_samples_per_second": 99.656,
|
| 278 |
+
"eval_steps_per_second": 49.828,
|
| 279 |
+
"step": 2600
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.125,
|
| 283 |
+
"eval_accuracy": 0.915625,
|
| 284 |
+
"eval_loss": 0.5321327447891235,
|
| 285 |
+
"eval_runtime": 3.0709,
|
| 286 |
+
"eval_samples_per_second": 104.205,
|
| 287 |
+
"eval_steps_per_second": 52.102,
|
| 288 |
+
"step": 2700
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.1666666666666667,
|
| 292 |
+
"eval_accuracy": 0.903125,
|
| 293 |
+
"eval_loss": 0.5952805280685425,
|
| 294 |
+
"eval_runtime": 2.8172,
|
| 295 |
+
"eval_samples_per_second": 113.588,
|
| 296 |
+
"eval_steps_per_second": 56.794,
|
| 297 |
+
"step": 2800
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.2083333333333333,
|
| 301 |
+
"eval_accuracy": 0.915625,
|
| 302 |
+
"eval_loss": 0.5122720003128052,
|
| 303 |
+
"eval_runtime": 2.9015,
|
| 304 |
+
"eval_samples_per_second": 110.288,
|
| 305 |
+
"eval_steps_per_second": 55.144,
|
| 306 |
+
"step": 2900
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.25,
|
| 310 |
+
"grad_norm": 0.010164022445678711,
|
| 311 |
+
"learning_rate": 2.750416666666667e-05,
|
| 312 |
+
"loss": 0.2205,
|
| 313 |
+
"step": 3000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.25,
|
| 317 |
+
"eval_accuracy": 0.884375,
|
| 318 |
+
"eval_loss": 0.7141415476799011,
|
| 319 |
+
"eval_runtime": 2.8765,
|
| 320 |
+
"eval_samples_per_second": 111.245,
|
| 321 |
+
"eval_steps_per_second": 55.622,
|
| 322 |
+
"step": 3000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.2916666666666667,
|
| 326 |
+
"eval_accuracy": 0.9,
|
| 327 |
+
"eval_loss": 0.5358812212944031,
|
| 328 |
+
"eval_runtime": 3.1535,
|
| 329 |
+
"eval_samples_per_second": 101.475,
|
| 330 |
+
"eval_steps_per_second": 50.737,
|
| 331 |
+
"step": 3100
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.3333333333333333,
|
| 335 |
+
"eval_accuracy": 0.915625,
|
| 336 |
+
"eval_loss": 0.5337697863578796,
|
| 337 |
+
"eval_runtime": 3.1817,
|
| 338 |
+
"eval_samples_per_second": 100.575,
|
| 339 |
+
"eval_steps_per_second": 50.287,
|
| 340 |
+
"step": 3200
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.375,
|
| 344 |
+
"eval_accuracy": 0.915625,
|
| 345 |
+
"eval_loss": 0.5300858020782471,
|
| 346 |
+
"eval_runtime": 2.8486,
|
| 347 |
+
"eval_samples_per_second": 112.336,
|
| 348 |
+
"eval_steps_per_second": 56.168,
|
| 349 |
+
"step": 3300
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.4166666666666667,
|
| 353 |
+
"eval_accuracy": 0.909375,
|
| 354 |
+
"eval_loss": 0.5486552119255066,
|
| 355 |
+
"eval_runtime": 2.9044,
|
| 356 |
+
"eval_samples_per_second": 110.177,
|
| 357 |
+
"eval_steps_per_second": 55.089,
|
| 358 |
+
"step": 3400
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.4583333333333333,
|
| 362 |
+
"grad_norm": 0.026199543848633766,
|
| 363 |
+
"learning_rate": 2.5420833333333336e-05,
|
| 364 |
+
"loss": 0.2262,
|
| 365 |
+
"step": 3500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.4583333333333333,
|
| 369 |
+
"eval_accuracy": 0.921875,
|
| 370 |
+
"eval_loss": 0.4780457615852356,
|
| 371 |
+
"eval_runtime": 3.1987,
|
| 372 |
+
"eval_samples_per_second": 100.041,
|
| 373 |
+
"eval_steps_per_second": 50.02,
|
| 374 |
+
"step": 3500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.5,
|
| 378 |
+
"eval_accuracy": 0.934375,
|
| 379 |
+
"eval_loss": 0.44137653708457947,
|
| 380 |
+
"eval_runtime": 3.1132,
|
| 381 |
+
"eval_samples_per_second": 102.788,
|
| 382 |
+
"eval_steps_per_second": 51.394,
|
| 383 |
+
"step": 3600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.5416666666666665,
|
| 387 |
+
"eval_accuracy": 0.91875,
|
| 388 |
+
"eval_loss": 0.5386074781417847,
|
| 389 |
+
"eval_runtime": 3.0032,
|
| 390 |
+
"eval_samples_per_second": 106.553,
|
| 391 |
+
"eval_steps_per_second": 53.276,
|
| 392 |
+
"step": 3700
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.5833333333333335,
|
| 396 |
+
"eval_accuracy": 0.940625,
|
| 397 |
+
"eval_loss": 0.3621799945831299,
|
| 398 |
+
"eval_runtime": 3.2102,
|
| 399 |
+
"eval_samples_per_second": 99.683,
|
| 400 |
+
"eval_steps_per_second": 49.842,
|
| 401 |
+
"step": 3800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.625,
|
| 405 |
+
"eval_accuracy": 0.9375,
|
| 406 |
+
"eval_loss": 0.3915068507194519,
|
| 407 |
+
"eval_runtime": 3.093,
|
| 408 |
+
"eval_samples_per_second": 103.459,
|
| 409 |
+
"eval_steps_per_second": 51.73,
|
| 410 |
+
"step": 3900
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.6666666666666665,
|
| 414 |
+
"grad_norm": 0.08724602311849594,
|
| 415 |
+
"learning_rate": 2.3337500000000005e-05,
|
| 416 |
+
"loss": 0.2279,
|
| 417 |
+
"step": 4000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.6666666666666665,
|
| 421 |
+
"eval_accuracy": 0.90625,
|
| 422 |
+
"eval_loss": 0.6075131297111511,
|
| 423 |
+
"eval_runtime": 2.8425,
|
| 424 |
+
"eval_samples_per_second": 112.577,
|
| 425 |
+
"eval_steps_per_second": 56.289,
|
| 426 |
+
"step": 4000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.7083333333333335,
|
| 430 |
+
"eval_accuracy": 0.934375,
|
| 431 |
+
"eval_loss": 0.40478381514549255,
|
| 432 |
+
"eval_runtime": 3.2632,
|
| 433 |
+
"eval_samples_per_second": 98.063,
|
| 434 |
+
"eval_steps_per_second": 49.032,
|
| 435 |
+
"step": 4100
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.75,
|
| 439 |
+
"eval_accuracy": 0.93125,
|
| 440 |
+
"eval_loss": 0.46917039155960083,
|
| 441 |
+
"eval_runtime": 2.851,
|
| 442 |
+
"eval_samples_per_second": 112.241,
|
| 443 |
+
"eval_steps_per_second": 56.12,
|
| 444 |
+
"step": 4200
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.7916666666666665,
|
| 448 |
+
"eval_accuracy": 0.93125,
|
| 449 |
+
"eval_loss": 0.4513327181339264,
|
| 450 |
+
"eval_runtime": 3.0309,
|
| 451 |
+
"eval_samples_per_second": 105.579,
|
| 452 |
+
"eval_steps_per_second": 52.79,
|
| 453 |
+
"step": 4300
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.8333333333333335,
|
| 457 |
+
"eval_accuracy": 0.934375,
|
| 458 |
+
"eval_loss": 0.4247393012046814,
|
| 459 |
+
"eval_runtime": 2.8817,
|
| 460 |
+
"eval_samples_per_second": 111.047,
|
| 461 |
+
"eval_steps_per_second": 55.523,
|
| 462 |
+
"step": 4400
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.875,
|
| 466 |
+
"grad_norm": 0.036841508001089096,
|
| 467 |
+
"learning_rate": 2.125416666666667e-05,
|
| 468 |
+
"loss": 0.1966,
|
| 469 |
+
"step": 4500
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.875,
|
| 473 |
+
"eval_accuracy": 0.925,
|
| 474 |
+
"eval_loss": 0.45489102602005005,
|
| 475 |
+
"eval_runtime": 2.8644,
|
| 476 |
+
"eval_samples_per_second": 111.718,
|
| 477 |
+
"eval_steps_per_second": 55.859,
|
| 478 |
+
"step": 4500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.9166666666666665,
|
| 482 |
+
"eval_accuracy": 0.925,
|
| 483 |
+
"eval_loss": 0.4184115529060364,
|
| 484 |
+
"eval_runtime": 2.9493,
|
| 485 |
+
"eval_samples_per_second": 108.499,
|
| 486 |
+
"eval_steps_per_second": 54.249,
|
| 487 |
+
"step": 4600
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.9583333333333335,
|
| 491 |
+
"eval_accuracy": 0.928125,
|
| 492 |
+
"eval_loss": 0.4278348982334137,
|
| 493 |
+
"eval_runtime": 3.348,
|
| 494 |
+
"eval_samples_per_second": 95.58,
|
| 495 |
+
"eval_steps_per_second": 47.79,
|
| 496 |
+
"step": 4700
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.0,
|
| 500 |
+
"eval_accuracy": 0.928125,
|
| 501 |
+
"eval_loss": 0.4810876250267029,
|
| 502 |
+
"eval_runtime": 3.1155,
|
| 503 |
+
"eval_samples_per_second": 102.711,
|
| 504 |
+
"eval_steps_per_second": 51.356,
|
| 505 |
+
"step": 4800
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 2.0416666666666665,
|
| 509 |
+
"eval_accuracy": 0.9375,
|
| 510 |
+
"eval_loss": 0.4345361590385437,
|
| 511 |
+
"eval_runtime": 2.957,
|
| 512 |
+
"eval_samples_per_second": 108.218,
|
| 513 |
+
"eval_steps_per_second": 54.109,
|
| 514 |
+
"step": 4900
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 2.0833333333333335,
|
| 518 |
+
"grad_norm": 0.011266867630183697,
|
| 519 |
+
"learning_rate": 1.9170833333333333e-05,
|
| 520 |
+
"loss": 0.1527,
|
| 521 |
+
"step": 5000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 2.0833333333333335,
|
| 525 |
+
"eval_accuracy": 0.9375,
|
| 526 |
+
"eval_loss": 0.43794384598731995,
|
| 527 |
+
"eval_runtime": 3.075,
|
| 528 |
+
"eval_samples_per_second": 104.065,
|
| 529 |
+
"eval_steps_per_second": 52.033,
|
| 530 |
+
"step": 5000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 2.125,
|
| 534 |
+
"eval_accuracy": 0.940625,
|
| 535 |
+
"eval_loss": 0.3865642249584198,
|
| 536 |
+
"eval_runtime": 2.8996,
|
| 537 |
+
"eval_samples_per_second": 110.359,
|
| 538 |
+
"eval_steps_per_second": 55.179,
|
| 539 |
+
"step": 5100
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.1666666666666665,
|
| 543 |
+
"eval_accuracy": 0.928125,
|
| 544 |
+
"eval_loss": 0.46689701080322266,
|
| 545 |
+
"eval_runtime": 3.0386,
|
| 546 |
+
"eval_samples_per_second": 105.311,
|
| 547 |
+
"eval_steps_per_second": 52.655,
|
| 548 |
+
"step": 5200
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.2083333333333335,
|
| 552 |
+
"eval_accuracy": 0.928125,
|
| 553 |
+
"eval_loss": 0.47042807936668396,
|
| 554 |
+
"eval_runtime": 2.9494,
|
| 555 |
+
"eval_samples_per_second": 108.497,
|
| 556 |
+
"eval_steps_per_second": 54.248,
|
| 557 |
+
"step": 5300
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.25,
|
| 561 |
+
"eval_accuracy": 0.93125,
|
| 562 |
+
"eval_loss": 0.4977450370788574,
|
| 563 |
+
"eval_runtime": 2.9827,
|
| 564 |
+
"eval_samples_per_second": 107.286,
|
| 565 |
+
"eval_steps_per_second": 53.643,
|
| 566 |
+
"step": 5400
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 2.2916666666666665,
|
| 570 |
+
"grad_norm": 0.013015041127800941,
|
| 571 |
+
"learning_rate": 1.7087500000000002e-05,
|
| 572 |
+
"loss": 0.1118,
|
| 573 |
+
"step": 5500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.2916666666666665,
|
| 577 |
+
"eval_accuracy": 0.93125,
|
| 578 |
+
"eval_loss": 0.48614591360092163,
|
| 579 |
+
"eval_runtime": 3.1215,
|
| 580 |
+
"eval_samples_per_second": 102.516,
|
| 581 |
+
"eval_steps_per_second": 51.258,
|
| 582 |
+
"step": 5500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.3333333333333335,
|
| 586 |
+
"eval_accuracy": 0.915625,
|
| 587 |
+
"eval_loss": 0.5790837407112122,
|
| 588 |
+
"eval_runtime": 3.1434,
|
| 589 |
+
"eval_samples_per_second": 101.799,
|
| 590 |
+
"eval_steps_per_second": 50.9,
|
| 591 |
+
"step": 5600
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.375,
|
| 595 |
+
"eval_accuracy": 0.93125,
|
| 596 |
+
"eval_loss": 0.466531366109848,
|
| 597 |
+
"eval_runtime": 3.1192,
|
| 598 |
+
"eval_samples_per_second": 102.59,
|
| 599 |
+
"eval_steps_per_second": 51.295,
|
| 600 |
+
"step": 5700
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.4166666666666665,
|
| 604 |
+
"eval_accuracy": 0.93125,
|
| 605 |
+
"eval_loss": 0.5104468464851379,
|
| 606 |
+
"eval_runtime": 2.9371,
|
| 607 |
+
"eval_samples_per_second": 108.952,
|
| 608 |
+
"eval_steps_per_second": 54.476,
|
| 609 |
+
"step": 5800
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.4583333333333335,
|
| 613 |
+
"eval_accuracy": 0.93125,
|
| 614 |
+
"eval_loss": 0.45083746314048767,
|
| 615 |
+
"eval_runtime": 3.069,
|
| 616 |
+
"eval_samples_per_second": 104.269,
|
| 617 |
+
"eval_steps_per_second": 52.135,
|
| 618 |
+
"step": 5900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.5,
|
| 622 |
+
"grad_norm": 0.011204572394490242,
|
| 623 |
+
"learning_rate": 1.5004166666666668e-05,
|
| 624 |
+
"loss": 0.0877,
|
| 625 |
+
"step": 6000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.5,
|
| 629 |
+
"eval_accuracy": 0.921875,
|
| 630 |
+
"eval_loss": 0.5052152872085571,
|
| 631 |
+
"eval_runtime": 3.164,
|
| 632 |
+
"eval_samples_per_second": 101.138,
|
| 633 |
+
"eval_steps_per_second": 50.569,
|
| 634 |
+
"step": 6000
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.5416666666666665,
|
| 638 |
+
"eval_accuracy": 0.9375,
|
| 639 |
+
"eval_loss": 0.4744489789009094,
|
| 640 |
+
"eval_runtime": 3.1285,
|
| 641 |
+
"eval_samples_per_second": 102.285,
|
| 642 |
+
"eval_steps_per_second": 51.142,
|
| 643 |
+
"step": 6100
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 2.5833333333333335,
|
| 647 |
+
"eval_accuracy": 0.93125,
|
| 648 |
+
"eval_loss": 0.5007426142692566,
|
| 649 |
+
"eval_runtime": 2.9779,
|
| 650 |
+
"eval_samples_per_second": 107.457,
|
| 651 |
+
"eval_steps_per_second": 53.729,
|
| 652 |
+
"step": 6200
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 2.625,
|
| 656 |
+
"eval_accuracy": 0.91875,
|
| 657 |
+
"eval_loss": 0.5636513829231262,
|
| 658 |
+
"eval_runtime": 2.8533,
|
| 659 |
+
"eval_samples_per_second": 112.152,
|
| 660 |
+
"eval_steps_per_second": 56.076,
|
| 661 |
+
"step": 6300
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.6666666666666665,
|
| 665 |
+
"eval_accuracy": 0.921875,
|
| 666 |
+
"eval_loss": 0.5763371586799622,
|
| 667 |
+
"eval_runtime": 2.9395,
|
| 668 |
+
"eval_samples_per_second": 108.862,
|
| 669 |
+
"eval_steps_per_second": 54.431,
|
| 670 |
+
"step": 6400
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.7083333333333335,
|
| 674 |
+
"grad_norm": 0.00426144152879715,
|
| 675 |
+
"learning_rate": 1.2920833333333333e-05,
|
| 676 |
+
"loss": 0.0661,
|
| 677 |
+
"step": 6500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 2.7083333333333335,
|
| 681 |
+
"eval_accuracy": 0.915625,
|
| 682 |
+
"eval_loss": 0.612991213798523,
|
| 683 |
+
"eval_runtime": 3.1905,
|
| 684 |
+
"eval_samples_per_second": 100.299,
|
| 685 |
+
"eval_steps_per_second": 50.15,
|
| 686 |
+
"step": 6500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.75,
|
| 690 |
+
"eval_accuracy": 0.9125,
|
| 691 |
+
"eval_loss": 0.5975683331489563,
|
| 692 |
+
"eval_runtime": 3.1067,
|
| 693 |
+
"eval_samples_per_second": 103.003,
|
| 694 |
+
"eval_steps_per_second": 51.502,
|
| 695 |
+
"step": 6600
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.7916666666666665,
|
| 699 |
+
"eval_accuracy": 0.91875,
|
| 700 |
+
"eval_loss": 0.5823736786842346,
|
| 701 |
+
"eval_runtime": 3.0985,
|
| 702 |
+
"eval_samples_per_second": 103.277,
|
| 703 |
+
"eval_steps_per_second": 51.639,
|
| 704 |
+
"step": 6700
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.8333333333333335,
|
| 708 |
+
"eval_accuracy": 0.915625,
|
| 709 |
+
"eval_loss": 0.6537708044052124,
|
| 710 |
+
"eval_runtime": 2.9363,
|
| 711 |
+
"eval_samples_per_second": 108.982,
|
| 712 |
+
"eval_steps_per_second": 54.491,
|
| 713 |
+
"step": 6800
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.875,
|
| 717 |
+
"eval_accuracy": 0.91875,
|
| 718 |
+
"eval_loss": 0.644100546836853,
|
| 719 |
+
"eval_runtime": 2.9441,
|
| 720 |
+
"eval_samples_per_second": 108.691,
|
| 721 |
+
"eval_steps_per_second": 54.346,
|
| 722 |
+
"step": 6900
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 2.9166666666666665,
|
| 726 |
+
"grad_norm": 0.007682121824473143,
|
| 727 |
+
"learning_rate": 1.08375e-05,
|
| 728 |
+
"loss": 0.0603,
|
| 729 |
+
"step": 7000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.9166666666666665,
|
| 733 |
+
"eval_accuracy": 0.91875,
|
| 734 |
+
"eval_loss": 0.6029552221298218,
|
| 735 |
+
"eval_runtime": 3.2097,
|
| 736 |
+
"eval_samples_per_second": 99.699,
|
| 737 |
+
"eval_steps_per_second": 49.849,
|
| 738 |
+
"step": 7000
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.9583333333333335,
|
| 742 |
+
"eval_accuracy": 0.925,
|
| 743 |
+
"eval_loss": 0.592232346534729,
|
| 744 |
+
"eval_runtime": 3.2116,
|
| 745 |
+
"eval_samples_per_second": 99.64,
|
| 746 |
+
"eval_steps_per_second": 49.82,
|
| 747 |
+
"step": 7100
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.0,
|
| 751 |
+
"eval_accuracy": 0.928125,
|
| 752 |
+
"eval_loss": 0.5717152953147888,
|
| 753 |
+
"eval_runtime": 3.1195,
|
| 754 |
+
"eval_samples_per_second": 102.581,
|
| 755 |
+
"eval_steps_per_second": 51.29,
|
| 756 |
+
"step": 7200
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 3.0416666666666665,
|
| 760 |
+
"eval_accuracy": 0.928125,
|
| 761 |
+
"eval_loss": 0.5890038013458252,
|
| 762 |
+
"eval_runtime": 3.0502,
|
| 763 |
+
"eval_samples_per_second": 104.91,
|
| 764 |
+
"eval_steps_per_second": 52.455,
|
| 765 |
+
"step": 7300
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 3.0833333333333335,
|
| 769 |
+
"eval_accuracy": 0.928125,
|
| 770 |
+
"eval_loss": 0.5868945121765137,
|
| 771 |
+
"eval_runtime": 2.9636,
|
| 772 |
+
"eval_samples_per_second": 107.977,
|
| 773 |
+
"eval_steps_per_second": 53.989,
|
| 774 |
+
"step": 7400
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.125,
|
| 778 |
+
"grad_norm": 0.004167845472693443,
|
| 779 |
+
"learning_rate": 8.754166666666667e-06,
|
| 780 |
+
"loss": 0.0321,
|
| 781 |
+
"step": 7500
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.125,
|
| 785 |
+
"eval_accuracy": 0.925,
|
| 786 |
+
"eval_loss": 0.5492711067199707,
|
| 787 |
+
"eval_runtime": 3.189,
|
| 788 |
+
"eval_samples_per_second": 100.346,
|
| 789 |
+
"eval_steps_per_second": 50.173,
|
| 790 |
+
"step": 7500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 3.1666666666666665,
|
| 794 |
+
"eval_accuracy": 0.925,
|
| 795 |
+
"eval_loss": 0.5748615264892578,
|
| 796 |
+
"eval_runtime": 3.1804,
|
| 797 |
+
"eval_samples_per_second": 100.615,
|
| 798 |
+
"eval_steps_per_second": 50.307,
|
| 799 |
+
"step": 7600
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 3.2083333333333335,
|
| 803 |
+
"eval_accuracy": 0.928125,
|
| 804 |
+
"eval_loss": 0.5854516625404358,
|
| 805 |
+
"eval_runtime": 3.2041,
|
| 806 |
+
"eval_samples_per_second": 99.873,
|
| 807 |
+
"eval_steps_per_second": 49.937,
|
| 808 |
+
"step": 7700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 3.25,
|
| 812 |
+
"eval_accuracy": 0.925,
|
| 813 |
+
"eval_loss": 0.5961952209472656,
|
| 814 |
+
"eval_runtime": 2.9283,
|
| 815 |
+
"eval_samples_per_second": 109.279,
|
| 816 |
+
"eval_steps_per_second": 54.639,
|
| 817 |
+
"step": 7800
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.2916666666666665,
|
| 821 |
+
"eval_accuracy": 0.925,
|
| 822 |
+
"eval_loss": 0.5780685544013977,
|
| 823 |
+
"eval_runtime": 3.0209,
|
| 824 |
+
"eval_samples_per_second": 105.929,
|
| 825 |
+
"eval_steps_per_second": 52.965,
|
| 826 |
+
"step": 7900
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 3.3333333333333335,
|
| 830 |
+
"grad_norm": 0.003411153331398964,
|
| 831 |
+
"learning_rate": 6.670833333333334e-06,
|
| 832 |
+
"loss": 0.019,
|
| 833 |
+
"step": 8000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.3333333333333335,
|
| 837 |
+
"eval_accuracy": 0.91875,
|
| 838 |
+
"eval_loss": 0.6126095652580261,
|
| 839 |
+
"eval_runtime": 3.2507,
|
| 840 |
+
"eval_samples_per_second": 98.44,
|
| 841 |
+
"eval_steps_per_second": 49.22,
|
| 842 |
+
"step": 8000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 3.375,
|
| 846 |
+
"eval_accuracy": 0.921875,
|
| 847 |
+
"eval_loss": 0.5971342325210571,
|
| 848 |
+
"eval_runtime": 3.2196,
|
| 849 |
+
"eval_samples_per_second": 99.39,
|
| 850 |
+
"eval_steps_per_second": 49.695,
|
| 851 |
+
"step": 8100
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.4166666666666665,
|
| 855 |
+
"eval_accuracy": 0.925,
|
| 856 |
+
"eval_loss": 0.5772821307182312,
|
| 857 |
+
"eval_runtime": 3.2295,
|
| 858 |
+
"eval_samples_per_second": 99.087,
|
| 859 |
+
"eval_steps_per_second": 49.544,
|
| 860 |
+
"step": 8200
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 3.4583333333333335,
|
| 864 |
+
"eval_accuracy": 0.921875,
|
| 865 |
+
"eval_loss": 0.5771257281303406,
|
| 866 |
+
"eval_runtime": 2.953,
|
| 867 |
+
"eval_samples_per_second": 108.366,
|
| 868 |
+
"eval_steps_per_second": 54.183,
|
| 869 |
+
"step": 8300
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 3.5,
|
| 873 |
+
"eval_accuracy": 0.91875,
|
| 874 |
+
"eval_loss": 0.59356290102005,
|
| 875 |
+
"eval_runtime": 3.1448,
|
| 876 |
+
"eval_samples_per_second": 101.756,
|
| 877 |
+
"eval_steps_per_second": 50.878,
|
| 878 |
+
"step": 8400
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 3.5416666666666665,
|
| 882 |
+
"grad_norm": 0.0046829418279230595,
|
| 883 |
+
"learning_rate": 4.5875000000000005e-06,
|
| 884 |
+
"loss": 0.046,
|
| 885 |
+
"step": 8500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.5416666666666665,
|
| 889 |
+
"eval_accuracy": 0.925,
|
| 890 |
+
"eval_loss": 0.5789353251457214,
|
| 891 |
+
"eval_runtime": 3.1737,
|
| 892 |
+
"eval_samples_per_second": 100.829,
|
| 893 |
+
"eval_steps_per_second": 50.414,
|
| 894 |
+
"step": 8500
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 3.5833333333333335,
|
| 898 |
+
"eval_accuracy": 0.921875,
|
| 899 |
+
"eval_loss": 0.5971996188163757,
|
| 900 |
+
"eval_runtime": 2.9558,
|
| 901 |
+
"eval_samples_per_second": 108.26,
|
| 902 |
+
"eval_steps_per_second": 54.13,
|
| 903 |
+
"step": 8600
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 3.625,
|
| 907 |
+
"eval_accuracy": 0.921875,
|
| 908 |
+
"eval_loss": 0.6019812822341919,
|
| 909 |
+
"eval_runtime": 2.9566,
|
| 910 |
+
"eval_samples_per_second": 108.231,
|
| 911 |
+
"eval_steps_per_second": 54.115,
|
| 912 |
+
"step": 8700
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 3.6666666666666665,
|
| 916 |
+
"eval_accuracy": 0.91875,
|
| 917 |
+
"eval_loss": 0.6330882906913757,
|
| 918 |
+
"eval_runtime": 2.9561,
|
| 919 |
+
"eval_samples_per_second": 108.253,
|
| 920 |
+
"eval_steps_per_second": 54.126,
|
| 921 |
+
"step": 8800
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 3.7083333333333335,
|
| 925 |
+
"eval_accuracy": 0.921875,
|
| 926 |
+
"eval_loss": 0.6137792468070984,
|
| 927 |
+
"eval_runtime": 3.2613,
|
| 928 |
+
"eval_samples_per_second": 98.121,
|
| 929 |
+
"eval_steps_per_second": 49.06,
|
| 930 |
+
"step": 8900
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 3.75,
|
| 934 |
+
"grad_norm": 0.004567369818687439,
|
| 935 |
+
"learning_rate": 2.504166666666667e-06,
|
| 936 |
+
"loss": 0.0324,
|
| 937 |
+
"step": 9000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 3.75,
|
| 941 |
+
"eval_accuracy": 0.921875,
|
| 942 |
+
"eval_loss": 0.6137566566467285,
|
| 943 |
+
"eval_runtime": 3.1314,
|
| 944 |
+
"eval_samples_per_second": 102.189,
|
| 945 |
+
"eval_steps_per_second": 51.095,
|
| 946 |
+
"step": 9000
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 3.7916666666666665,
|
| 950 |
+
"eval_accuracy": 0.925,
|
| 951 |
+
"eval_loss": 0.5908929109573364,
|
| 952 |
+
"eval_runtime": 2.9973,
|
| 953 |
+
"eval_samples_per_second": 106.764,
|
| 954 |
+
"eval_steps_per_second": 53.382,
|
| 955 |
+
"step": 9100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 3.8333333333333335,
|
| 959 |
+
"eval_accuracy": 0.925,
|
| 960 |
+
"eval_loss": 0.5938516855239868,
|
| 961 |
+
"eval_runtime": 2.9851,
|
| 962 |
+
"eval_samples_per_second": 107.201,
|
| 963 |
+
"eval_steps_per_second": 53.6,
|
| 964 |
+
"step": 9200
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 3.875,
|
| 968 |
+
"eval_accuracy": 0.925,
|
| 969 |
+
"eval_loss": 0.5955865383148193,
|
| 970 |
+
"eval_runtime": 2.996,
|
| 971 |
+
"eval_samples_per_second": 106.808,
|
| 972 |
+
"eval_steps_per_second": 53.404,
|
| 973 |
+
"step": 9300
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 3.9166666666666665,
|
| 977 |
+
"eval_accuracy": 0.925,
|
| 978 |
+
"eval_loss": 0.5968046188354492,
|
| 979 |
+
"eval_runtime": 3.1695,
|
| 980 |
+
"eval_samples_per_second": 100.964,
|
| 981 |
+
"eval_steps_per_second": 50.482,
|
| 982 |
+
"step": 9400
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 3.9583333333333335,
|
| 986 |
+
"grad_norm": 0.004296157043427229,
|
| 987 |
+
"learning_rate": 4.2083333333333336e-07,
|
| 988 |
+
"loss": 0.0303,
|
| 989 |
+
"step": 9500
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 3.9583333333333335,
|
| 993 |
+
"eval_accuracy": 0.925,
|
| 994 |
+
"eval_loss": 0.5974746942520142,
|
| 995 |
+
"eval_runtime": 3.105,
|
| 996 |
+
"eval_samples_per_second": 103.058,
|
| 997 |
+
"eval_steps_per_second": 51.529,
|
| 998 |
+
"step": 9500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 4.0,
|
| 1002 |
+
"eval_accuracy": 0.925,
|
| 1003 |
+
"eval_loss": 0.5987403988838196,
|
| 1004 |
+
"eval_runtime": 2.9896,
|
| 1005 |
+
"eval_samples_per_second": 107.039,
|
| 1006 |
+
"eval_steps_per_second": 53.519,
|
| 1007 |
+
"step": 9600
|
| 1008 |
+
}
|
| 1009 |
+
],
|
| 1010 |
+
"logging_steps": 500,
|
| 1011 |
+
"max_steps": 9600,
|
| 1012 |
+
"num_input_tokens_seen": 0,
|
| 1013 |
+
"num_train_epochs": 4,
|
| 1014 |
+
"save_steps": 100,
|
| 1015 |
+
"stateful_callbacks": {
|
| 1016 |
+
"TrainerControl": {
|
| 1017 |
+
"args": {
|
| 1018 |
+
"should_epoch_stop": false,
|
| 1019 |
+
"should_evaluate": false,
|
| 1020 |
+
"should_log": false,
|
| 1021 |
+
"should_save": true,
|
| 1022 |
+
"should_training_stop": true
|
| 1023 |
+
},
|
| 1024 |
+
"attributes": {}
|
| 1025 |
+
}
|
| 1026 |
+
},
|
| 1027 |
+
"total_flos": 1328331579648360.0,
|
| 1028 |
+
"train_batch_size": 2,
|
| 1029 |
+
"trial_name": null,
|
| 1030 |
+
"trial_params": null
|
| 1031 |
+
}
|
checkpoints/checkpoint-9600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e993d463aba6969e9f6ddada547c66913d99677754065e5d3d85907f9219bf4
|
| 3 |
+
size 5304
|
checkpoints/checkpoint-9600/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
final_model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3495ebbeb992d3657676a010b5254965845fc80006e6952e42fc74c66e79dcef
|
| 3 |
+
size 438017556
|
final_model_meta.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"model_name": "google-bert/bert-base-uncased"}
|
test_outputs.jsonl
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"step": 100, "loss": 0.4754180908203125, "accuracy": 0.89375, "runtime": 2.8838, "samples_per_second": 110.967, "steps_per_second": 55.483}
|
| 2 |
+
{"step": 200, "loss": 0.7783083319664001, "accuracy": 0.84375, "runtime": 2.8722, "samples_per_second": 111.413, "steps_per_second": 55.706}
|
| 3 |
+
{"step": 300, "loss": 0.4844598174095154, "accuracy": 0.91875, "runtime": 2.9679, "samples_per_second": 107.821, "steps_per_second": 53.911}
|
| 4 |
+
{"step": 400, "loss": 0.5653066635131836, "accuracy": 0.8875, "runtime": 2.8114, "samples_per_second": 113.822, "steps_per_second": 56.911}
|
| 5 |
+
{"step": 500, "loss": 0.4834604263305664, "accuracy": 0.89375, "runtime": 3.0663, "samples_per_second": 104.361, "steps_per_second": 52.18}
|
| 6 |
+
{"step": 600, "loss": 0.3776414394378662, "accuracy": 0.928125, "runtime": 2.9683, "samples_per_second": 107.806, "steps_per_second": 53.903}
|
| 7 |
+
{"step": 700, "loss": 0.7465925216674805, "accuracy": 0.846875, "runtime": 3.1091, "samples_per_second": 102.924, "steps_per_second": 51.462}
|
| 8 |
+
{"step": 800, "loss": 0.43123188614845276, "accuracy": 0.91875, "runtime": 2.9983, "samples_per_second": 106.727, "steps_per_second": 53.363}
|
| 9 |
+
{"step": 900, "loss": 0.3499329686164856, "accuracy": 0.921875, "runtime": 3.1929, "samples_per_second": 100.224, "steps_per_second": 50.112}
|
| 10 |
+
{"step": 1000, "loss": 0.7541207671165466, "accuracy": 0.834375, "runtime": 2.9939, "samples_per_second": 106.885, "steps_per_second": 53.442}
|
| 11 |
+
{"step": 1100, "loss": 0.47846031188964844, "accuracy": 0.909375, "runtime": 2.965, "samples_per_second": 107.926, "steps_per_second": 53.963}
|
| 12 |
+
{"step": 1200, "loss": 0.6214954257011414, "accuracy": 0.903125, "runtime": 2.9259, "samples_per_second": 109.368, "steps_per_second": 54.684}
|
| 13 |
+
{"step": 1300, "loss": 0.42658787965774536, "accuracy": 0.915625, "runtime": 3.0223, "samples_per_second": 105.879, "steps_per_second": 52.939}
|
| 14 |
+
{"step": 1400, "loss": 0.3985453248023987, "accuracy": 0.90625, "runtime": 2.9117, "samples_per_second": 109.903, "steps_per_second": 54.952}
|
| 15 |
+
{"step": 1500, "loss": 0.5395756363868713, "accuracy": 0.9, "runtime": 2.8898, "samples_per_second": 110.733, "steps_per_second": 55.366}
|
| 16 |
+
{"step": 1600, "loss": 0.48287898302078247, "accuracy": 0.90625, "runtime": 3.07, "samples_per_second": 104.235, "steps_per_second": 52.118}
|
| 17 |
+
{"step": 1700, "loss": 0.4366416037082672, "accuracy": 0.915625, "runtime": 3.0743, "samples_per_second": 104.09, "steps_per_second": 52.045}
|
| 18 |
+
{"step": 1800, "loss": 0.5095136761665344, "accuracy": 0.9125, "runtime": 3.0831, "samples_per_second": 103.792, "steps_per_second": 51.896}
|
| 19 |
+
{"step": 1900, "loss": 0.4809819161891937, "accuracy": 0.915625, "runtime": 2.8721, "samples_per_second": 111.416, "steps_per_second": 55.708}
|
| 20 |
+
{"step": 2000, "loss": 0.6681545376777649, "accuracy": 0.8875, "runtime": 2.8368, "samples_per_second": 112.803, "steps_per_second": 56.402}
|
| 21 |
+
{"step": 2100, "loss": 0.4883766770362854, "accuracy": 0.9125, "runtime": 3.2206, "samples_per_second": 99.36, "steps_per_second": 49.68}
|
| 22 |
+
{"step": 2200, "loss": 0.37708765268325806, "accuracy": 0.915625, "runtime": 3.2053, "samples_per_second": 99.835, "steps_per_second": 49.917}
|
| 23 |
+
{"step": 2300, "loss": 0.38682788610458374, "accuracy": 0.91875, "runtime": 2.8559, "samples_per_second": 112.048, "steps_per_second": 56.024}
|
| 24 |
+
{"step": 2400, "loss": 0.5085960626602173, "accuracy": 0.90625, "runtime": 2.9762, "samples_per_second": 107.518, "steps_per_second": 53.759}
|
| 25 |
+
{"step": 2500, "loss": 0.467485249042511, "accuracy": 0.921875, "runtime": 2.8829, "samples_per_second": 111.001, "steps_per_second": 55.5}
|
| 26 |
+
{"step": 2600, "loss": 0.6294034719467163, "accuracy": 0.89375, "runtime": 3.211, "samples_per_second": 99.656, "steps_per_second": 49.828}
|
| 27 |
+
{"step": 2700, "loss": 0.5321327447891235, "accuracy": 0.915625, "runtime": 3.0709, "samples_per_second": 104.205, "steps_per_second": 52.102}
|
| 28 |
+
{"step": 2800, "loss": 0.5952805280685425, "accuracy": 0.903125, "runtime": 2.8172, "samples_per_second": 113.588, "steps_per_second": 56.794}
|
| 29 |
+
{"step": 2900, "loss": 0.5122720003128052, "accuracy": 0.915625, "runtime": 2.9015, "samples_per_second": 110.288, "steps_per_second": 55.144}
|
| 30 |
+
{"step": 3000, "loss": 0.7141415476799011, "accuracy": 0.884375, "runtime": 2.8765, "samples_per_second": 111.245, "steps_per_second": 55.622}
|
| 31 |
+
{"step": 3100, "loss": 0.5358812212944031, "accuracy": 0.9, "runtime": 3.1535, "samples_per_second": 101.475, "steps_per_second": 50.737}
|
| 32 |
+
{"step": 3200, "loss": 0.5337697863578796, "accuracy": 0.915625, "runtime": 3.1817, "samples_per_second": 100.575, "steps_per_second": 50.287}
|
| 33 |
+
{"step": 3300, "loss": 0.5300858020782471, "accuracy": 0.915625, "runtime": 2.8486, "samples_per_second": 112.336, "steps_per_second": 56.168}
|
| 34 |
+
{"step": 3400, "loss": 0.5486552119255066, "accuracy": 0.909375, "runtime": 2.9044, "samples_per_second": 110.177, "steps_per_second": 55.089}
|
| 35 |
+
{"step": 3500, "loss": 0.4780457615852356, "accuracy": 0.921875, "runtime": 3.1987, "samples_per_second": 100.041, "steps_per_second": 50.02}
|
| 36 |
+
{"step": 3600, "loss": 0.44137653708457947, "accuracy": 0.934375, "runtime": 3.1132, "samples_per_second": 102.788, "steps_per_second": 51.394}
|
| 37 |
+
{"step": 3700, "loss": 0.5386074781417847, "accuracy": 0.91875, "runtime": 3.0032, "samples_per_second": 106.553, "steps_per_second": 53.276}
|
| 38 |
+
{"step": 3800, "loss": 0.3621799945831299, "accuracy": 0.940625, "runtime": 3.2102, "samples_per_second": 99.683, "steps_per_second": 49.842}
|
| 39 |
+
{"step": 3900, "loss": 0.3915068507194519, "accuracy": 0.9375, "runtime": 3.093, "samples_per_second": 103.459, "steps_per_second": 51.73}
|
| 40 |
+
{"step": 4000, "loss": 0.6075131297111511, "accuracy": 0.90625, "runtime": 2.8425, "samples_per_second": 112.577, "steps_per_second": 56.289}
|
| 41 |
+
{"step": 4100, "loss": 0.40478381514549255, "accuracy": 0.934375, "runtime": 3.2632, "samples_per_second": 98.063, "steps_per_second": 49.032}
|
| 42 |
+
{"step": 4200, "loss": 0.46917039155960083, "accuracy": 0.93125, "runtime": 2.851, "samples_per_second": 112.241, "steps_per_second": 56.12}
|
| 43 |
+
{"step": 4300, "loss": 0.4513327181339264, "accuracy": 0.93125, "runtime": 3.0309, "samples_per_second": 105.579, "steps_per_second": 52.79}
|
| 44 |
+
{"step": 4400, "loss": 0.4247393012046814, "accuracy": 0.934375, "runtime": 2.8817, "samples_per_second": 111.047, "steps_per_second": 55.523}
|
| 45 |
+
{"step": 4500, "loss": 0.45489102602005005, "accuracy": 0.925, "runtime": 2.8644, "samples_per_second": 111.718, "steps_per_second": 55.859}
|
| 46 |
+
{"step": 4600, "loss": 0.4184115529060364, "accuracy": 0.925, "runtime": 2.9493, "samples_per_second": 108.499, "steps_per_second": 54.249}
|
| 47 |
+
{"step": 4700, "loss": 0.4278348982334137, "accuracy": 0.928125, "runtime": 3.348, "samples_per_second": 95.58, "steps_per_second": 47.79}
|
| 48 |
+
{"step": 4800, "loss": 0.4810876250267029, "accuracy": 0.928125, "runtime": 3.1155, "samples_per_second": 102.711, "steps_per_second": 51.356}
|
| 49 |
+
{"step": 4900, "loss": 0.4345361590385437, "accuracy": 0.9375, "runtime": 2.957, "samples_per_second": 108.218, "steps_per_second": 54.109}
|
| 50 |
+
{"step": 5000, "loss": 0.43794384598731995, "accuracy": 0.9375, "runtime": 3.075, "samples_per_second": 104.065, "steps_per_second": 52.033}
|
| 51 |
+
{"step": 5100, "loss": 0.3865642249584198, "accuracy": 0.940625, "runtime": 2.8996, "samples_per_second": 110.359, "steps_per_second": 55.179}
|
| 52 |
+
{"step": 5200, "loss": 0.46689701080322266, "accuracy": 0.928125, "runtime": 3.0386, "samples_per_second": 105.311, "steps_per_second": 52.655}
|
| 53 |
+
{"step": 5300, "loss": 0.47042807936668396, "accuracy": 0.928125, "runtime": 2.9494, "samples_per_second": 108.497, "steps_per_second": 54.248}
|
| 54 |
+
{"step": 5400, "loss": 0.4977450370788574, "accuracy": 0.93125, "runtime": 2.9827, "samples_per_second": 107.286, "steps_per_second": 53.643}
|
| 55 |
+
{"step": 5500, "loss": 0.48614591360092163, "accuracy": 0.93125, "runtime": 3.1215, "samples_per_second": 102.516, "steps_per_second": 51.258}
|
| 56 |
+
{"step": 5600, "loss": 0.5790837407112122, "accuracy": 0.915625, "runtime": 3.1434, "samples_per_second": 101.799, "steps_per_second": 50.9}
|
| 57 |
+
{"step": 5700, "loss": 0.466531366109848, "accuracy": 0.93125, "runtime": 3.1192, "samples_per_second": 102.59, "steps_per_second": 51.295}
|
| 58 |
+
{"step": 5800, "loss": 0.5104468464851379, "accuracy": 0.93125, "runtime": 2.9371, "samples_per_second": 108.952, "steps_per_second": 54.476}
|
| 59 |
+
{"step": 5900, "loss": 0.45083746314048767, "accuracy": 0.93125, "runtime": 3.069, "samples_per_second": 104.269, "steps_per_second": 52.135}
|
| 60 |
+
{"step": 6000, "loss": 0.5052152872085571, "accuracy": 0.921875, "runtime": 3.164, "samples_per_second": 101.138, "steps_per_second": 50.569}
|
| 61 |
+
{"step": 6100, "loss": 0.4744489789009094, "accuracy": 0.9375, "runtime": 3.1285, "samples_per_second": 102.285, "steps_per_second": 51.142}
|
| 62 |
+
{"step": 6200, "loss": 0.5007426142692566, "accuracy": 0.93125, "runtime": 2.9779, "samples_per_second": 107.457, "steps_per_second": 53.729}
|
| 63 |
+
{"step": 6300, "loss": 0.5636513829231262, "accuracy": 0.91875, "runtime": 2.8533, "samples_per_second": 112.152, "steps_per_second": 56.076}
|
| 64 |
+
{"step": 6400, "loss": 0.5763371586799622, "accuracy": 0.921875, "runtime": 2.9395, "samples_per_second": 108.862, "steps_per_second": 54.431}
|
| 65 |
+
{"step": 6500, "loss": 0.612991213798523, "accuracy": 0.915625, "runtime": 3.1905, "samples_per_second": 100.299, "steps_per_second": 50.15}
|
| 66 |
+
{"step": 6600, "loss": 0.5975683331489563, "accuracy": 0.9125, "runtime": 3.1067, "samples_per_second": 103.003, "steps_per_second": 51.502}
|
| 67 |
+
{"step": 6700, "loss": 0.5823736786842346, "accuracy": 0.91875, "runtime": 3.0985, "samples_per_second": 103.277, "steps_per_second": 51.639}
|
| 68 |
+
{"step": 6800, "loss": 0.6537708044052124, "accuracy": 0.915625, "runtime": 2.9363, "samples_per_second": 108.982, "steps_per_second": 54.491}
|
| 69 |
+
{"step": 6900, "loss": 0.644100546836853, "accuracy": 0.91875, "runtime": 2.9441, "samples_per_second": 108.691, "steps_per_second": 54.346}
|
| 70 |
+
{"step": 7000, "loss": 0.6029552221298218, "accuracy": 0.91875, "runtime": 3.2097, "samples_per_second": 99.699, "steps_per_second": 49.849}
|
| 71 |
+
{"step": 7100, "loss": 0.592232346534729, "accuracy": 0.925, "runtime": 3.2116, "samples_per_second": 99.64, "steps_per_second": 49.82}
|
| 72 |
+
{"step": 7200, "loss": 0.5717152953147888, "accuracy": 0.928125, "runtime": 3.1195, "samples_per_second": 102.581, "steps_per_second": 51.29}
|
| 73 |
+
{"step": 7300, "loss": 0.5890038013458252, "accuracy": 0.928125, "runtime": 3.0502, "samples_per_second": 104.91, "steps_per_second": 52.455}
|
| 74 |
+
{"step": 7400, "loss": 0.5868945121765137, "accuracy": 0.928125, "runtime": 2.9636, "samples_per_second": 107.977, "steps_per_second": 53.989}
|
| 75 |
+
{"step": 7500, "loss": 0.5492711067199707, "accuracy": 0.925, "runtime": 3.189, "samples_per_second": 100.346, "steps_per_second": 50.173}
|
| 76 |
+
{"step": 7600, "loss": 0.5748615264892578, "accuracy": 0.925, "runtime": 3.1804, "samples_per_second": 100.615, "steps_per_second": 50.307}
|
| 77 |
+
{"step": 7700, "loss": 0.5854516625404358, "accuracy": 0.928125, "runtime": 3.2041, "samples_per_second": 99.873, "steps_per_second": 49.937}
|
| 78 |
+
{"step": 7800, "loss": 0.5961952209472656, "accuracy": 0.925, "runtime": 2.9283, "samples_per_second": 109.279, "steps_per_second": 54.639}
|
| 79 |
+
{"step": 7900, "loss": 0.5780685544013977, "accuracy": 0.925, "runtime": 3.0209, "samples_per_second": 105.929, "steps_per_second": 52.965}
|
| 80 |
+
{"step": 8000, "loss": 0.6126095652580261, "accuracy": 0.91875, "runtime": 3.2507, "samples_per_second": 98.44, "steps_per_second": 49.22}
|
| 81 |
+
{"step": 8100, "loss": 0.5971342325210571, "accuracy": 0.921875, "runtime": 3.2196, "samples_per_second": 99.39, "steps_per_second": 49.695}
|
| 82 |
+
{"step": 8200, "loss": 0.5772821307182312, "accuracy": 0.925, "runtime": 3.2295, "samples_per_second": 99.087, "steps_per_second": 49.544}
|
| 83 |
+
{"step": 8300, "loss": 0.5771257281303406, "accuracy": 0.921875, "runtime": 2.953, "samples_per_second": 108.366, "steps_per_second": 54.183}
|
| 84 |
+
{"step": 8400, "loss": 0.59356290102005, "accuracy": 0.91875, "runtime": 3.1448, "samples_per_second": 101.756, "steps_per_second": 50.878}
|
| 85 |
+
{"step": 8500, "loss": 0.5789353251457214, "accuracy": 0.925, "runtime": 3.1737, "samples_per_second": 100.829, "steps_per_second": 50.414}
|
| 86 |
+
{"step": 8600, "loss": 0.5971996188163757, "accuracy": 0.921875, "runtime": 2.9558, "samples_per_second": 108.26, "steps_per_second": 54.13}
|
| 87 |
+
{"step": 8700, "loss": 0.6019812822341919, "accuracy": 0.921875, "runtime": 2.9566, "samples_per_second": 108.231, "steps_per_second": 54.115}
|
| 88 |
+
{"step": 8800, "loss": 0.6330882906913757, "accuracy": 0.91875, "runtime": 2.9561, "samples_per_second": 108.253, "steps_per_second": 54.126}
|
| 89 |
+
{"step": 8900, "loss": 0.6137792468070984, "accuracy": 0.921875, "runtime": 3.2613, "samples_per_second": 98.121, "steps_per_second": 49.06}
|
| 90 |
+
{"step": 9000, "loss": 0.6137566566467285, "accuracy": 0.921875, "runtime": 3.1314, "samples_per_second": 102.189, "steps_per_second": 51.095}
|
| 91 |
+
{"step": 9100, "loss": 0.5908929109573364, "accuracy": 0.925, "runtime": 2.9973, "samples_per_second": 106.764, "steps_per_second": 53.382}
|
| 92 |
+
{"step": 9200, "loss": 0.5938516855239868, "accuracy": 0.925, "runtime": 2.9851, "samples_per_second": 107.201, "steps_per_second": 53.6}
|
| 93 |
+
{"step": 9300, "loss": 0.5955865383148193, "accuracy": 0.925, "runtime": 2.996, "samples_per_second": 106.808, "steps_per_second": 53.404}
|
| 94 |
+
{"step": 9400, "loss": 0.5968046188354492, "accuracy": 0.925, "runtime": 3.1695, "samples_per_second": 100.964, "steps_per_second": 50.482}
|
| 95 |
+
{"step": 9500, "loss": 0.5974746942520142, "accuracy": 0.925, "runtime": 3.105, "samples_per_second": 103.058, "steps_per_second": 51.529}
|
| 96 |
+
{"step": 9600, "loss": 0.5987403988838196, "accuracy": 0.925, "runtime": 2.9896, "samples_per_second": 107.039, "steps_per_second": 53.519}
|