Upload 7 files
Browse files- config.json +128 -0
- preprocessor_config.json +9 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +913 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,128 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "soba1911/xlsr-wav2vec2",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForSpeechClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "sum",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.1,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "wav2vec2_clf",
|
| 57 |
+
"gradient_checkpointing": false,
|
| 58 |
+
"hidden_act": "gelu",
|
| 59 |
+
"hidden_dropout": 0.1,
|
| 60 |
+
"hidden_size": 1024,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "bonafide",
|
| 63 |
+
"1": "spoof"
|
| 64 |
+
},
|
| 65 |
+
"initializer_range": 0.02,
|
| 66 |
+
"intermediate_size": 4096,
|
| 67 |
+
"label2id": {
|
| 68 |
+
"bonafide": 0,
|
| 69 |
+
"spoof": 1
|
| 70 |
+
},
|
| 71 |
+
"layer_norm_eps": 1e-05,
|
| 72 |
+
"layerdrop": 0.1,
|
| 73 |
+
"mask_channel_length": 10,
|
| 74 |
+
"mask_channel_min_space": 1,
|
| 75 |
+
"mask_channel_other": 0.0,
|
| 76 |
+
"mask_channel_prob": 0.0,
|
| 77 |
+
"mask_channel_selection": "static",
|
| 78 |
+
"mask_feature_length": 10,
|
| 79 |
+
"mask_feature_min_masks": 0,
|
| 80 |
+
"mask_feature_prob": 0.0,
|
| 81 |
+
"mask_time_length": 10,
|
| 82 |
+
"mask_time_min_masks": 2,
|
| 83 |
+
"mask_time_min_space": 1,
|
| 84 |
+
"mask_time_other": 0.0,
|
| 85 |
+
"mask_time_prob": 0.075,
|
| 86 |
+
"mask_time_selection": "static",
|
| 87 |
+
"model_type": "wav2vec2",
|
| 88 |
+
"num_adapter_layers": 3,
|
| 89 |
+
"num_attention_heads": 16,
|
| 90 |
+
"num_codevector_groups": 2,
|
| 91 |
+
"num_codevectors_per_group": 320,
|
| 92 |
+
"num_conv_pos_embedding_groups": 16,
|
| 93 |
+
"num_conv_pos_embeddings": 128,
|
| 94 |
+
"num_feat_extract_layers": 7,
|
| 95 |
+
"num_hidden_layers": 24,
|
| 96 |
+
"num_negatives": 100,
|
| 97 |
+
"output_hidden_size": 1024,
|
| 98 |
+
"pad_token_id": 0,
|
| 99 |
+
"pooling_mode": "mean",
|
| 100 |
+
"problem_type": "single_label_classification",
|
| 101 |
+
"proj_codevector_dim": 768,
|
| 102 |
+
"tdnn_dilation": [
|
| 103 |
+
1,
|
| 104 |
+
2,
|
| 105 |
+
3,
|
| 106 |
+
1,
|
| 107 |
+
1
|
| 108 |
+
],
|
| 109 |
+
"tdnn_dim": [
|
| 110 |
+
512,
|
| 111 |
+
512,
|
| 112 |
+
512,
|
| 113 |
+
512,
|
| 114 |
+
1500
|
| 115 |
+
],
|
| 116 |
+
"tdnn_kernel": [
|
| 117 |
+
5,
|
| 118 |
+
3,
|
| 119 |
+
3,
|
| 120 |
+
1,
|
| 121 |
+
1
|
| 122 |
+
],
|
| 123 |
+
"torch_dtype": "float32",
|
| 124 |
+
"transformers_version": "4.42.0.dev0",
|
| 125 |
+
"use_weighted_layer_sum": false,
|
| 126 |
+
"vocab_size": 32,
|
| 127 |
+
"xvector_output_dim": 512
|
| 128 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a4479abe1652b884f1d94f8ac69ebd859e6918177f93457a3444f03259ed265
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4aec5c2f056acaa2fbffcd9c5be9cf23e3440cbb2644f199cc6997e40426b30f
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,913 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.23143277929728592,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 550,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004207868714496108,
|
| 13 |
+
"grad_norm": 10.497783660888672,
|
| 14 |
+
"learning_rate": 0.0004909090909090909,
|
| 15 |
+
"loss": 0.7938,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004207868714496108,
|
| 20 |
+
"eval_accuracy": 0.5304018259048462,
|
| 21 |
+
"eval_loss": 0.7331877946853638,
|
| 22 |
+
"eval_runtime": 568.2697,
|
| 23 |
+
"eval_samples_per_second": 8.364,
|
| 24 |
+
"eval_steps_per_second": 2.092,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.008415737428992216,
|
| 29 |
+
"grad_norm": 2.557694435119629,
|
| 30 |
+
"learning_rate": 0.00048181818181818184,
|
| 31 |
+
"loss": 0.7504,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.008415737428992216,
|
| 36 |
+
"eval_accuracy": 0.5634336471557617,
|
| 37 |
+
"eval_loss": 0.7359612584114075,
|
| 38 |
+
"eval_runtime": 559.3032,
|
| 39 |
+
"eval_samples_per_second": 8.498,
|
| 40 |
+
"eval_steps_per_second": 2.126,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.012623606143488323,
|
| 45 |
+
"grad_norm": 3.9564919471740723,
|
| 46 |
+
"learning_rate": 0.0004727272727272727,
|
| 47 |
+
"loss": 0.7866,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.012623606143488323,
|
| 52 |
+
"eval_accuracy": 0.4085840582847595,
|
| 53 |
+
"eval_loss": 0.7127184271812439,
|
| 54 |
+
"eval_runtime": 567.0848,
|
| 55 |
+
"eval_samples_per_second": 8.381,
|
| 56 |
+
"eval_steps_per_second": 2.097,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.016831474857984433,
|
| 61 |
+
"grad_norm": 29.89333724975586,
|
| 62 |
+
"learning_rate": 0.00046363636363636366,
|
| 63 |
+
"loss": 0.678,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.016831474857984433,
|
| 68 |
+
"eval_accuracy": 0.6875657439231873,
|
| 69 |
+
"eval_loss": 0.6557931303977966,
|
| 70 |
+
"eval_runtime": 563.9541,
|
| 71 |
+
"eval_samples_per_second": 8.428,
|
| 72 |
+
"eval_steps_per_second": 2.108,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.021039343572480537,
|
| 77 |
+
"grad_norm": 1.9482946395874023,
|
| 78 |
+
"learning_rate": 0.00045454545454545455,
|
| 79 |
+
"loss": 0.8634,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.021039343572480537,
|
| 84 |
+
"eval_accuracy": 0.6572691202163696,
|
| 85 |
+
"eval_loss": 0.613047182559967,
|
| 86 |
+
"eval_runtime": 565.9198,
|
| 87 |
+
"eval_samples_per_second": 8.399,
|
| 88 |
+
"eval_steps_per_second": 2.101,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.025247212286976645,
|
| 93 |
+
"grad_norm": 1.0520250797271729,
|
| 94 |
+
"learning_rate": 0.00044545454545454543,
|
| 95 |
+
"loss": 0.5117,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.025247212286976645,
|
| 100 |
+
"eval_accuracy": 0.7416368722915649,
|
| 101 |
+
"eval_loss": 0.5154983997344971,
|
| 102 |
+
"eval_runtime": 566.7191,
|
| 103 |
+
"eval_samples_per_second": 8.387,
|
| 104 |
+
"eval_steps_per_second": 2.098,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.029455081001472753,
|
| 109 |
+
"grad_norm": 3.958573341369629,
|
| 110 |
+
"learning_rate": 0.00043636363636363637,
|
| 111 |
+
"loss": 0.6066,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.029455081001472753,
|
| 116 |
+
"eval_accuracy": 0.7858194708824158,
|
| 117 |
+
"eval_loss": 0.5255141854286194,
|
| 118 |
+
"eval_runtime": 561.6163,
|
| 119 |
+
"eval_samples_per_second": 8.463,
|
| 120 |
+
"eval_steps_per_second": 2.117,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.033662949715968865,
|
| 125 |
+
"grad_norm": 7.255526542663574,
|
| 126 |
+
"learning_rate": 0.00042727272727272726,
|
| 127 |
+
"loss": 0.5282,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.033662949715968865,
|
| 132 |
+
"eval_accuracy": 0.8045445084571838,
|
| 133 |
+
"eval_loss": 0.6335883140563965,
|
| 134 |
+
"eval_runtime": 565.1826,
|
| 135 |
+
"eval_samples_per_second": 8.41,
|
| 136 |
+
"eval_steps_per_second": 2.104,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.03787081843046497,
|
| 141 |
+
"grad_norm": 4.439815521240234,
|
| 142 |
+
"learning_rate": 0.00041818181818181814,
|
| 143 |
+
"loss": 0.3694,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.03787081843046497,
|
| 148 |
+
"eval_accuracy": 0.8161161541938782,
|
| 149 |
+
"eval_loss": 0.464495450258255,
|
| 150 |
+
"eval_runtime": 562.0768,
|
| 151 |
+
"eval_samples_per_second": 8.456,
|
| 152 |
+
"eval_steps_per_second": 2.115,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.042078687144961074,
|
| 157 |
+
"grad_norm": 1.75038743019104,
|
| 158 |
+
"learning_rate": 0.00040909090909090913,
|
| 159 |
+
"loss": 0.4786,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.042078687144961074,
|
| 164 |
+
"eval_accuracy": 0.8079107999801636,
|
| 165 |
+
"eval_loss": 0.4171961545944214,
|
| 166 |
+
"eval_runtime": 555.7896,
|
| 167 |
+
"eval_samples_per_second": 8.552,
|
| 168 |
+
"eval_steps_per_second": 2.139,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.046286555859457186,
|
| 173 |
+
"grad_norm": 2.398599624633789,
|
| 174 |
+
"learning_rate": 0.0004,
|
| 175 |
+
"loss": 0.3394,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.046286555859457186,
|
| 180 |
+
"eval_accuracy": 0.8739743232727051,
|
| 181 |
+
"eval_loss": 0.353726327419281,
|
| 182 |
+
"eval_runtime": 558.2722,
|
| 183 |
+
"eval_samples_per_second": 8.514,
|
| 184 |
+
"eval_steps_per_second": 2.13,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.05049442457395329,
|
| 189 |
+
"grad_norm": 2.4667985439300537,
|
| 190 |
+
"learning_rate": 0.00039090909090909096,
|
| 191 |
+
"loss": 0.4313,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.05049442457395329,
|
| 196 |
+
"eval_accuracy": 0.8889122605323792,
|
| 197 |
+
"eval_loss": 0.2820029556751251,
|
| 198 |
+
"eval_runtime": 563.3341,
|
| 199 |
+
"eval_samples_per_second": 8.437,
|
| 200 |
+
"eval_steps_per_second": 2.111,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.0547022932884494,
|
| 205 |
+
"grad_norm": 1.1293903589248657,
|
| 206 |
+
"learning_rate": 0.00038181818181818184,
|
| 207 |
+
"loss": 0.3352,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.0547022932884494,
|
| 212 |
+
"eval_accuracy": 0.8889122605323792,
|
| 213 |
+
"eval_loss": 0.28811973333358765,
|
| 214 |
+
"eval_runtime": 563.2561,
|
| 215 |
+
"eval_samples_per_second": 8.438,
|
| 216 |
+
"eval_steps_per_second": 2.111,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.05891016200294551,
|
| 221 |
+
"grad_norm": 17.52765655517578,
|
| 222 |
+
"learning_rate": 0.00037272727272727273,
|
| 223 |
+
"loss": 0.3772,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.05891016200294551,
|
| 228 |
+
"eval_accuracy": 0.8116979002952576,
|
| 229 |
+
"eval_loss": 0.4701159596443176,
|
| 230 |
+
"eval_runtime": 565.5511,
|
| 231 |
+
"eval_samples_per_second": 8.404,
|
| 232 |
+
"eval_steps_per_second": 2.102,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06311803071744161,
|
| 237 |
+
"grad_norm": 2.1096954345703125,
|
| 238 |
+
"learning_rate": 0.00036363636363636367,
|
| 239 |
+
"loss": 0.3699,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.06311803071744161,
|
| 244 |
+
"eval_accuracy": 0.8693456649780273,
|
| 245 |
+
"eval_loss": 0.32162848114967346,
|
| 246 |
+
"eval_runtime": 561.7642,
|
| 247 |
+
"eval_samples_per_second": 8.461,
|
| 248 |
+
"eval_steps_per_second": 2.117,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.06732589943193773,
|
| 253 |
+
"grad_norm": 0.7017167210578918,
|
| 254 |
+
"learning_rate": 0.00035454545454545455,
|
| 255 |
+
"loss": 0.4895,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.06732589943193773,
|
| 260 |
+
"eval_accuracy": 0.8817588686943054,
|
| 261 |
+
"eval_loss": 0.36469992995262146,
|
| 262 |
+
"eval_runtime": 563.6361,
|
| 263 |
+
"eval_samples_per_second": 8.433,
|
| 264 |
+
"eval_steps_per_second": 2.11,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.07153376814643383,
|
| 269 |
+
"grad_norm": 0.48568668961524963,
|
| 270 |
+
"learning_rate": 0.00034545454545454544,
|
| 271 |
+
"loss": 0.46,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.07153376814643383,
|
| 276 |
+
"eval_accuracy": 0.8449400663375854,
|
| 277 |
+
"eval_loss": 0.35531896352767944,
|
| 278 |
+
"eval_runtime": 566.3467,
|
| 279 |
+
"eval_samples_per_second": 8.392,
|
| 280 |
+
"eval_steps_per_second": 2.099,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.07574163686092994,
|
| 285 |
+
"grad_norm": 0.6846579909324646,
|
| 286 |
+
"learning_rate": 0.0003363636363636364,
|
| 287 |
+
"loss": 0.3931,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.07574163686092994,
|
| 292 |
+
"eval_accuracy": 0.8916473984718323,
|
| 293 |
+
"eval_loss": 0.2812640070915222,
|
| 294 |
+
"eval_runtime": 565.8446,
|
| 295 |
+
"eval_samples_per_second": 8.4,
|
| 296 |
+
"eval_steps_per_second": 2.101,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.07994950557542604,
|
| 301 |
+
"grad_norm": 2.391395092010498,
|
| 302 |
+
"learning_rate": 0.00032727272727272726,
|
| 303 |
+
"loss": 0.3666,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.07994950557542604,
|
| 308 |
+
"eval_accuracy": 0.8321060538291931,
|
| 309 |
+
"eval_loss": 0.4278256595134735,
|
| 310 |
+
"eval_runtime": 562.6059,
|
| 311 |
+
"eval_samples_per_second": 8.448,
|
| 312 |
+
"eval_steps_per_second": 2.113,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08415737428992215,
|
| 317 |
+
"grad_norm": 1.708675503730774,
|
| 318 |
+
"learning_rate": 0.0003181818181818182,
|
| 319 |
+
"loss": 0.3471,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.08415737428992215,
|
| 324 |
+
"eval_accuracy": 0.9156322479248047,
|
| 325 |
+
"eval_loss": 0.23789770901203156,
|
| 326 |
+
"eval_runtime": 565.4642,
|
| 327 |
+
"eval_samples_per_second": 8.405,
|
| 328 |
+
"eval_steps_per_second": 2.103,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.08836524300441827,
|
| 333 |
+
"grad_norm": 0.4107959270477295,
|
| 334 |
+
"learning_rate": 0.0003090909090909091,
|
| 335 |
+
"loss": 0.4351,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.08836524300441827,
|
| 340 |
+
"eval_accuracy": 0.8889122605323792,
|
| 341 |
+
"eval_loss": 0.29446446895599365,
|
| 342 |
+
"eval_runtime": 562.2329,
|
| 343 |
+
"eval_samples_per_second": 8.454,
|
| 344 |
+
"eval_steps_per_second": 2.115,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.09257311171891437,
|
| 349 |
+
"grad_norm": 10.296160697937012,
|
| 350 |
+
"learning_rate": 0.0003,
|
| 351 |
+
"loss": 0.262,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.09257311171891437,
|
| 356 |
+
"eval_accuracy": 0.9017462730407715,
|
| 357 |
+
"eval_loss": 0.307280033826828,
|
| 358 |
+
"eval_runtime": 561.3167,
|
| 359 |
+
"eval_samples_per_second": 8.468,
|
| 360 |
+
"eval_steps_per_second": 2.118,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.09678098043341048,
|
| 365 |
+
"grad_norm": 0.6356618404388428,
|
| 366 |
+
"learning_rate": 0.0002909090909090909,
|
| 367 |
+
"loss": 0.3291,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.09678098043341048,
|
| 372 |
+
"eval_accuracy": 0.8752366900444031,
|
| 373 |
+
"eval_loss": 0.3042367994785309,
|
| 374 |
+
"eval_runtime": 567.4141,
|
| 375 |
+
"eval_samples_per_second": 8.377,
|
| 376 |
+
"eval_steps_per_second": 2.095,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.10098884914790658,
|
| 381 |
+
"grad_norm": 1.1508342027664185,
|
| 382 |
+
"learning_rate": 0.0002818181818181818,
|
| 383 |
+
"loss": 0.239,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.10098884914790658,
|
| 388 |
+
"eval_accuracy": 0.9150010347366333,
|
| 389 |
+
"eval_loss": 0.20924389362335205,
|
| 390 |
+
"eval_runtime": 564.144,
|
| 391 |
+
"eval_samples_per_second": 8.425,
|
| 392 |
+
"eval_steps_per_second": 2.108,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.1051967178624027,
|
| 397 |
+
"grad_norm": 1.0988367795944214,
|
| 398 |
+
"learning_rate": 0.00027272727272727274,
|
| 399 |
+
"loss": 0.4306,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.1051967178624027,
|
| 404 |
+
"eval_accuracy": 0.905954122543335,
|
| 405 |
+
"eval_loss": 0.24196282029151917,
|
| 406 |
+
"eval_runtime": 565.5038,
|
| 407 |
+
"eval_samples_per_second": 8.405,
|
| 408 |
+
"eval_steps_per_second": 2.103,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1094045865768988,
|
| 413 |
+
"grad_norm": 1.5871909856796265,
|
| 414 |
+
"learning_rate": 0.0002636363636363636,
|
| 415 |
+
"loss": 0.2757,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1094045865768988,
|
| 420 |
+
"eval_accuracy": 0.8415737152099609,
|
| 421 |
+
"eval_loss": 0.3900565505027771,
|
| 422 |
+
"eval_runtime": 566.4852,
|
| 423 |
+
"eval_samples_per_second": 8.39,
|
| 424 |
+
"eval_steps_per_second": 2.099,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.11361245529139491,
|
| 429 |
+
"grad_norm": 0.33946001529693604,
|
| 430 |
+
"learning_rate": 0.0002545454545454545,
|
| 431 |
+
"loss": 0.1851,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.11361245529139491,
|
| 436 |
+
"eval_accuracy": 0.9040606021881104,
|
| 437 |
+
"eval_loss": 0.2729853093624115,
|
| 438 |
+
"eval_runtime": 563.3502,
|
| 439 |
+
"eval_samples_per_second": 8.437,
|
| 440 |
+
"eval_steps_per_second": 2.111,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.11782032400589101,
|
| 445 |
+
"grad_norm": 0.7806110978126526,
|
| 446 |
+
"learning_rate": 0.00024545454545454545,
|
| 447 |
+
"loss": 0.2655,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.11782032400589101,
|
| 452 |
+
"eval_accuracy": 0.8645066022872925,
|
| 453 |
+
"eval_loss": 0.3478306829929352,
|
| 454 |
+
"eval_runtime": 565.4573,
|
| 455 |
+
"eval_samples_per_second": 8.406,
|
| 456 |
+
"eval_steps_per_second": 2.103,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.12202819272038712,
|
| 461 |
+
"grad_norm": 1.8319846391677856,
|
| 462 |
+
"learning_rate": 0.00023636363636363636,
|
| 463 |
+
"loss": 0.2119,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.12202819272038712,
|
| 468 |
+
"eval_accuracy": 0.9112139940261841,
|
| 469 |
+
"eval_loss": 0.23500923812389374,
|
| 470 |
+
"eval_runtime": 563.7148,
|
| 471 |
+
"eval_samples_per_second": 8.432,
|
| 472 |
+
"eval_steps_per_second": 2.109,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.12623606143488322,
|
| 477 |
+
"grad_norm": 1.5416840314865112,
|
| 478 |
+
"learning_rate": 0.00022727272727272727,
|
| 479 |
+
"loss": 0.1477,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.12623606143488322,
|
| 484 |
+
"eval_accuracy": 0.9219440221786499,
|
| 485 |
+
"eval_loss": 0.3015914261341095,
|
| 486 |
+
"eval_runtime": 561.7407,
|
| 487 |
+
"eval_samples_per_second": 8.461,
|
| 488 |
+
"eval_steps_per_second": 2.117,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.13044393014937933,
|
| 493 |
+
"grad_norm": 3.576409101486206,
|
| 494 |
+
"learning_rate": 0.00021818181818181818,
|
| 495 |
+
"loss": 0.2137,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.13044393014937933,
|
| 500 |
+
"eval_accuracy": 0.9282558560371399,
|
| 501 |
+
"eval_loss": 0.30129292607307434,
|
| 502 |
+
"eval_runtime": 563.5899,
|
| 503 |
+
"eval_samples_per_second": 8.433,
|
| 504 |
+
"eval_steps_per_second": 2.11,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.13465179886387546,
|
| 509 |
+
"grad_norm": 9.926487922668457,
|
| 510 |
+
"learning_rate": 0.00020909090909090907,
|
| 511 |
+
"loss": 0.1136,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.13465179886387546,
|
| 516 |
+
"eval_accuracy": 0.9307805299758911,
|
| 517 |
+
"eval_loss": 0.2483457773923874,
|
| 518 |
+
"eval_runtime": 564.5525,
|
| 519 |
+
"eval_samples_per_second": 8.419,
|
| 520 |
+
"eval_steps_per_second": 2.106,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.13885966757837157,
|
| 525 |
+
"grad_norm": 1.2627246379852295,
|
| 526 |
+
"learning_rate": 0.0002,
|
| 527 |
+
"loss": 0.197,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.13885966757837157,
|
| 532 |
+
"eval_accuracy": 0.913528323173523,
|
| 533 |
+
"eval_loss": 0.3319007456302643,
|
| 534 |
+
"eval_runtime": 557.94,
|
| 535 |
+
"eval_samples_per_second": 8.519,
|
| 536 |
+
"eval_steps_per_second": 2.131,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.14306753629286767,
|
| 541 |
+
"grad_norm": 14.595890998840332,
|
| 542 |
+
"learning_rate": 0.00019090909090909092,
|
| 543 |
+
"loss": 0.1492,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.14306753629286767,
|
| 548 |
+
"eval_accuracy": 0.8773406147956848,
|
| 549 |
+
"eval_loss": 0.4927380383014679,
|
| 550 |
+
"eval_runtime": 565.4025,
|
| 551 |
+
"eval_samples_per_second": 8.406,
|
| 552 |
+
"eval_steps_per_second": 2.103,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.14727540500736377,
|
| 557 |
+
"grad_norm": 0.08015768229961395,
|
| 558 |
+
"learning_rate": 0.00018181818181818183,
|
| 559 |
+
"loss": 0.0394,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.14727540500736377,
|
| 564 |
+
"eval_accuracy": 0.8889122605323792,
|
| 565 |
+
"eval_loss": 0.5241063833236694,
|
| 566 |
+
"eval_runtime": 567.4727,
|
| 567 |
+
"eval_samples_per_second": 8.376,
|
| 568 |
+
"eval_steps_per_second": 2.095,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.15148327372185988,
|
| 573 |
+
"grad_norm": 1.9829438924789429,
|
| 574 |
+
"learning_rate": 0.00017272727272727272,
|
| 575 |
+
"loss": 0.284,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.15148327372185988,
|
| 580 |
+
"eval_accuracy": 0.924679160118103,
|
| 581 |
+
"eval_loss": 0.3283034563064575,
|
| 582 |
+
"eval_runtime": 565.6368,
|
| 583 |
+
"eval_samples_per_second": 8.403,
|
| 584 |
+
"eval_steps_per_second": 2.102,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.15569114243635598,
|
| 589 |
+
"grad_norm": 0.26033639907836914,
|
| 590 |
+
"learning_rate": 0.00016363636363636363,
|
| 591 |
+
"loss": 0.1096,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.15569114243635598,
|
| 596 |
+
"eval_accuracy": 0.9179465770721436,
|
| 597 |
+
"eval_loss": 0.2719731330871582,
|
| 598 |
+
"eval_runtime": 568.9149,
|
| 599 |
+
"eval_samples_per_second": 8.355,
|
| 600 |
+
"eval_steps_per_second": 2.09,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.1598990111508521,
|
| 605 |
+
"grad_norm": 0.4041542112827301,
|
| 606 |
+
"learning_rate": 0.00015454545454545454,
|
| 607 |
+
"loss": 0.3147,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.1598990111508521,
|
| 612 |
+
"eval_accuracy": 0.9389858841896057,
|
| 613 |
+
"eval_loss": 0.20388783514499664,
|
| 614 |
+
"eval_runtime": 576.8924,
|
| 615 |
+
"eval_samples_per_second": 8.239,
|
| 616 |
+
"eval_steps_per_second": 2.061,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.1641068798653482,
|
| 621 |
+
"grad_norm": 0.7307087182998657,
|
| 622 |
+
"learning_rate": 0.00014545454545454546,
|
| 623 |
+
"loss": 0.1,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.1641068798653482,
|
| 628 |
+
"eval_accuracy": 0.9320428967475891,
|
| 629 |
+
"eval_loss": 0.2630753219127655,
|
| 630 |
+
"eval_runtime": 572.9749,
|
| 631 |
+
"eval_samples_per_second": 8.295,
|
| 632 |
+
"eval_steps_per_second": 2.075,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.1683147485798443,
|
| 637 |
+
"grad_norm": 0.9732351303100586,
|
| 638 |
+
"learning_rate": 0.00013636363636363637,
|
| 639 |
+
"loss": 0.1954,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.1683147485798443,
|
| 644 |
+
"eval_accuracy": 0.93309485912323,
|
| 645 |
+
"eval_loss": 0.265785813331604,
|
| 646 |
+
"eval_runtime": 571.5753,
|
| 647 |
+
"eval_samples_per_second": 8.316,
|
| 648 |
+
"eval_steps_per_second": 2.08,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.17252261729434043,
|
| 653 |
+
"grad_norm": 3.890988826751709,
|
| 654 |
+
"learning_rate": 0.00012727272727272725,
|
| 655 |
+
"loss": 0.1332,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.17252261729434043,
|
| 660 |
+
"eval_accuracy": 0.9469808340072632,
|
| 661 |
+
"eval_loss": 0.2041083574295044,
|
| 662 |
+
"eval_runtime": 572.7877,
|
| 663 |
+
"eval_samples_per_second": 8.298,
|
| 664 |
+
"eval_steps_per_second": 2.076,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.17673048600883653,
|
| 669 |
+
"grad_norm": 0.22084006667137146,
|
| 670 |
+
"learning_rate": 0.00011818181818181818,
|
| 671 |
+
"loss": 0.116,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.17673048600883653,
|
| 676 |
+
"eval_accuracy": 0.9162634015083313,
|
| 677 |
+
"eval_loss": 0.3249155879020691,
|
| 678 |
+
"eval_runtime": 572.8138,
|
| 679 |
+
"eval_samples_per_second": 8.298,
|
| 680 |
+
"eval_steps_per_second": 2.076,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.18093835472333264,
|
| 685 |
+
"grad_norm": 0.09070427715778351,
|
| 686 |
+
"learning_rate": 0.00010909090909090909,
|
| 687 |
+
"loss": 0.3625,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.18093835472333264,
|
| 692 |
+
"eval_accuracy": 0.9541342258453369,
|
| 693 |
+
"eval_loss": 0.16712911427021027,
|
| 694 |
+
"eval_runtime": 570.8705,
|
| 695 |
+
"eval_samples_per_second": 8.326,
|
| 696 |
+
"eval_steps_per_second": 2.083,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.18514622343782874,
|
| 701 |
+
"grad_norm": 3.719890594482422,
|
| 702 |
+
"learning_rate": 0.0001,
|
| 703 |
+
"loss": 0.203,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.18514622343782874,
|
| 708 |
+
"eval_accuracy": 0.9591836929321289,
|
| 709 |
+
"eval_loss": 0.11601703613996506,
|
| 710 |
+
"eval_runtime": 559.6205,
|
| 711 |
+
"eval_samples_per_second": 8.493,
|
| 712 |
+
"eval_steps_per_second": 2.125,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.18935409215232485,
|
| 717 |
+
"grad_norm": 1.8163635730743408,
|
| 718 |
+
"learning_rate": 9.090909090909092e-05,
|
| 719 |
+
"loss": 0.1065,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.18935409215232485,
|
| 724 |
+
"eval_accuracy": 0.9663370251655579,
|
| 725 |
+
"eval_loss": 0.11517694592475891,
|
| 726 |
+
"eval_runtime": 564.4747,
|
| 727 |
+
"eval_samples_per_second": 8.42,
|
| 728 |
+
"eval_steps_per_second": 2.106,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.19356196086682095,
|
| 733 |
+
"grad_norm": 0.489663302898407,
|
| 734 |
+
"learning_rate": 8.181818181818182e-05,
|
| 735 |
+
"loss": 0.0797,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.19356196086682095,
|
| 740 |
+
"eval_accuracy": 0.9524511098861694,
|
| 741 |
+
"eval_loss": 0.17924398183822632,
|
| 742 |
+
"eval_runtime": 566.5969,
|
| 743 |
+
"eval_samples_per_second": 8.389,
|
| 744 |
+
"eval_steps_per_second": 2.098,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.19776982958131706,
|
| 749 |
+
"grad_norm": 0.12014146894216537,
|
| 750 |
+
"learning_rate": 7.272727272727273e-05,
|
| 751 |
+
"loss": 0.158,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.19776982958131706,
|
| 756 |
+
"eval_accuracy": 0.9545550346374512,
|
| 757 |
+
"eval_loss": 0.17613738775253296,
|
| 758 |
+
"eval_runtime": 565.847,
|
| 759 |
+
"eval_samples_per_second": 8.4,
|
| 760 |
+
"eval_steps_per_second": 2.101,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.20197769829581316,
|
| 765 |
+
"grad_norm": 0.30841195583343506,
|
| 766 |
+
"learning_rate": 6.363636363636363e-05,
|
| 767 |
+
"loss": 0.1226,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.20197769829581316,
|
| 772 |
+
"eval_accuracy": 0.954344630241394,
|
| 773 |
+
"eval_loss": 0.1746922880411148,
|
| 774 |
+
"eval_runtime": 566.983,
|
| 775 |
+
"eval_samples_per_second": 8.383,
|
| 776 |
+
"eval_steps_per_second": 2.097,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.20618556701030927,
|
| 781 |
+
"grad_norm": 0.17474113404750824,
|
| 782 |
+
"learning_rate": 5.4545454545454546e-05,
|
| 783 |
+
"loss": 0.0502,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.20618556701030927,
|
| 788 |
+
"eval_accuracy": 0.9596044421195984,
|
| 789 |
+
"eval_loss": 0.16830527782440186,
|
| 790 |
+
"eval_runtime": 566.0043,
|
| 791 |
+
"eval_samples_per_second": 8.397,
|
| 792 |
+
"eval_steps_per_second": 2.101,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2103934357248054,
|
| 797 |
+
"grad_norm": 0.19428321719169617,
|
| 798 |
+
"learning_rate": 4.545454545454546e-05,
|
| 799 |
+
"loss": 0.0346,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2103934357248054,
|
| 804 |
+
"eval_accuracy": 0.9585524797439575,
|
| 805 |
+
"eval_loss": 0.17505739629268646,
|
| 806 |
+
"eval_runtime": 564.2226,
|
| 807 |
+
"eval_samples_per_second": 8.424,
|
| 808 |
+
"eval_steps_per_second": 2.107,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.2146013044393015,
|
| 813 |
+
"grad_norm": 0.13528193533420563,
|
| 814 |
+
"learning_rate": 3.6363636363636364e-05,
|
| 815 |
+
"loss": 0.0168,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.2146013044393015,
|
| 820 |
+
"eval_accuracy": 0.958131730556488,
|
| 821 |
+
"eval_loss": 0.18199646472930908,
|
| 822 |
+
"eval_runtime": 563.8378,
|
| 823 |
+
"eval_samples_per_second": 8.43,
|
| 824 |
+
"eval_steps_per_second": 2.109,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.2188091731537976,
|
| 829 |
+
"grad_norm": 0.03537715971469879,
|
| 830 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 831 |
+
"loss": 0.1428,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.2188091731537976,
|
| 836 |
+
"eval_accuracy": 0.9562381505966187,
|
| 837 |
+
"eval_loss": 0.1906786561012268,
|
| 838 |
+
"eval_runtime": 567.1515,
|
| 839 |
+
"eval_samples_per_second": 8.38,
|
| 840 |
+
"eval_steps_per_second": 2.096,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.22301704186829371,
|
| 845 |
+
"grad_norm": 0.5100229978561401,
|
| 846 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 847 |
+
"loss": 0.0336,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.22301704186829371,
|
| 852 |
+
"eval_accuracy": 0.9503471255302429,
|
| 853 |
+
"eval_loss": 0.2207876741886139,
|
| 854 |
+
"eval_runtime": 564.0513,
|
| 855 |
+
"eval_samples_per_second": 8.427,
|
| 856 |
+
"eval_steps_per_second": 2.108,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.22722491058278982,
|
| 861 |
+
"grad_norm": 5.639271259307861,
|
| 862 |
+
"learning_rate": 9.090909090909091e-06,
|
| 863 |
+
"loss": 0.1161,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.22722491058278982,
|
| 868 |
+
"eval_accuracy": 0.9488744139671326,
|
| 869 |
+
"eval_loss": 0.2279575914144516,
|
| 870 |
+
"eval_runtime": 564.6939,
|
| 871 |
+
"eval_samples_per_second": 8.417,
|
| 872 |
+
"eval_steps_per_second": 2.106,
|
| 873 |
+
"step": 540
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.23143277929728592,
|
| 877 |
+
"grad_norm": 0.09282880276441574,
|
| 878 |
+
"learning_rate": 0.0,
|
| 879 |
+
"loss": 0.0377,
|
| 880 |
+
"step": 550
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.23143277929728592,
|
| 884 |
+
"eval_accuracy": 0.9495055675506592,
|
| 885 |
+
"eval_loss": 0.22491338849067688,
|
| 886 |
+
"eval_runtime": 556.6035,
|
| 887 |
+
"eval_samples_per_second": 8.539,
|
| 888 |
+
"eval_steps_per_second": 2.136,
|
| 889 |
+
"step": 550
|
| 890 |
+
}
|
| 891 |
+
],
|
| 892 |
+
"logging_steps": 10,
|
| 893 |
+
"max_steps": 550,
|
| 894 |
+
"num_input_tokens_seen": 0,
|
| 895 |
+
"num_train_epochs": 1,
|
| 896 |
+
"save_steps": 10,
|
| 897 |
+
"stateful_callbacks": {
|
| 898 |
+
"TrainerControl": {
|
| 899 |
+
"args": {
|
| 900 |
+
"should_epoch_stop": false,
|
| 901 |
+
"should_evaluate": false,
|
| 902 |
+
"should_log": false,
|
| 903 |
+
"should_save": true,
|
| 904 |
+
"should_training_stop": true
|
| 905 |
+
},
|
| 906 |
+
"attributes": {}
|
| 907 |
+
}
|
| 908 |
+
},
|
| 909 |
+
"total_flos": 6.208452205931052e+17,
|
| 910 |
+
"train_batch_size": 4,
|
| 911 |
+
"trial_name": null,
|
| 912 |
+
"trial_params": null
|
| 913 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b22510bd5c536935aa1df4a61597c3aefca6b79bd8643603be4f076f18d9857
|
| 3 |
+
size 5112
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "'": 5, "-": 6, "a": 7, "b": 8, "c": 9, "d": 10, "e": 11, "f": 12, "g": 13, "h": 14, "i": 15, "j": 16, "k": 17, "l": 18, "m": 19, "n": 20, "o": 21, "p": 22, "q": 23, "r": 24, "s": 25, "t": 26, "u": 27, "v": 28, "w": 29, "x": 30, "y": 31, "z": 32}
|