Upload 6 files
Browse files- config.json +128 -0
- preprocessor_config.json +9 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +897 -0
- training_args.bin +3 -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": "4.41.0.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:d2b214d4edc827b1f4780776bb3c2c1c02fdbb6b3c53a752eb23889b0adf5317
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9fbacf1e3e2d01be4c10ce5f17fec9be9543287c3d7269f23f53f77939ab7588
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,897 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.24064171122994651,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 540,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004456327985739751,
|
| 13 |
+
"grad_norm": 1.5512940883636475,
|
| 14 |
+
"learning_rate": 9.818181818181818e-05,
|
| 15 |
+
"loss": 0.4724,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004456327985739751,
|
| 20 |
+
"eval_accuracy": 0.89683598279953,
|
| 21 |
+
"eval_loss": 0.34007999300956726,
|
| 22 |
+
"eval_runtime": 549.9948,
|
| 23 |
+
"eval_samples_per_second": 8.16,
|
| 24 |
+
"eval_steps_per_second": 2.04,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.008912655971479501,
|
| 29 |
+
"grad_norm": 1.4623041152954102,
|
| 30 |
+
"learning_rate": 9.636363636363637e-05,
|
| 31 |
+
"loss": 0.2715,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.008912655971479501,
|
| 36 |
+
"eval_accuracy": 0.89683598279953,
|
| 37 |
+
"eval_loss": 0.39523470401763916,
|
| 38 |
+
"eval_runtime": 543.4137,
|
| 39 |
+
"eval_samples_per_second": 8.259,
|
| 40 |
+
"eval_steps_per_second": 2.065,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.013368983957219251,
|
| 45 |
+
"grad_norm": 0.6407843828201294,
|
| 46 |
+
"learning_rate": 9.454545454545455e-05,
|
| 47 |
+
"loss": 0.3566,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.013368983957219251,
|
| 52 |
+
"eval_accuracy": 0.89683598279953,
|
| 53 |
+
"eval_loss": 0.38253363966941833,
|
| 54 |
+
"eval_runtime": 536.8913,
|
| 55 |
+
"eval_samples_per_second": 8.359,
|
| 56 |
+
"eval_steps_per_second": 2.09,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.017825311942959002,
|
| 61 |
+
"grad_norm": 1.9206137657165527,
|
| 62 |
+
"learning_rate": 9.272727272727273e-05,
|
| 63 |
+
"loss": 0.2892,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.017825311942959002,
|
| 68 |
+
"eval_accuracy": 0.89683598279953,
|
| 69 |
+
"eval_loss": 0.36841902136802673,
|
| 70 |
+
"eval_runtime": 539.4459,
|
| 71 |
+
"eval_samples_per_second": 8.32,
|
| 72 |
+
"eval_steps_per_second": 2.08,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.022281639928698752,
|
| 77 |
+
"grad_norm": 1.0926023721694946,
|
| 78 |
+
"learning_rate": 9.090909090909092e-05,
|
| 79 |
+
"loss": 0.3157,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.022281639928698752,
|
| 84 |
+
"eval_accuracy": 0.89683598279953,
|
| 85 |
+
"eval_loss": 0.3572486639022827,
|
| 86 |
+
"eval_runtime": 542.4121,
|
| 87 |
+
"eval_samples_per_second": 8.274,
|
| 88 |
+
"eval_steps_per_second": 2.069,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.026737967914438502,
|
| 93 |
+
"grad_norm": 3.161236524581909,
|
| 94 |
+
"learning_rate": 8.90909090909091e-05,
|
| 95 |
+
"loss": 0.3792,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.026737967914438502,
|
| 100 |
+
"eval_accuracy": 0.89683598279953,
|
| 101 |
+
"eval_loss": 0.3476051688194275,
|
| 102 |
+
"eval_runtime": 538.7398,
|
| 103 |
+
"eval_samples_per_second": 8.331,
|
| 104 |
+
"eval_steps_per_second": 2.083,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.031194295900178252,
|
| 109 |
+
"grad_norm": 1.0513850450515747,
|
| 110 |
+
"learning_rate": 8.727272727272727e-05,
|
| 111 |
+
"loss": 0.3938,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.031194295900178252,
|
| 116 |
+
"eval_accuracy": 0.89683598279953,
|
| 117 |
+
"eval_loss": 0.317058265209198,
|
| 118 |
+
"eval_runtime": 540.0617,
|
| 119 |
+
"eval_samples_per_second": 8.31,
|
| 120 |
+
"eval_steps_per_second": 2.078,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.035650623885918005,
|
| 125 |
+
"grad_norm": 5.6621479988098145,
|
| 126 |
+
"learning_rate": 8.545454545454545e-05,
|
| 127 |
+
"loss": 0.3962,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.035650623885918005,
|
| 132 |
+
"eval_accuracy": 0.89683598279953,
|
| 133 |
+
"eval_loss": 0.2908115088939667,
|
| 134 |
+
"eval_runtime": 541.5124,
|
| 135 |
+
"eval_samples_per_second": 8.288,
|
| 136 |
+
"eval_steps_per_second": 2.072,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.040106951871657755,
|
| 141 |
+
"grad_norm": 11.542706489562988,
|
| 142 |
+
"learning_rate": 8.363636363636364e-05,
|
| 143 |
+
"loss": 0.3536,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.040106951871657755,
|
| 148 |
+
"eval_accuracy": 0.89683598279953,
|
| 149 |
+
"eval_loss": 0.2727506160736084,
|
| 150 |
+
"eval_runtime": 542.0266,
|
| 151 |
+
"eval_samples_per_second": 8.28,
|
| 152 |
+
"eval_steps_per_second": 2.07,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.044563279857397504,
|
| 157 |
+
"grad_norm": 0.7968679070472717,
|
| 158 |
+
"learning_rate": 8.181818181818183e-05,
|
| 159 |
+
"loss": 0.2338,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.044563279857397504,
|
| 164 |
+
"eval_accuracy": 0.898172914981842,
|
| 165 |
+
"eval_loss": 0.21359723806381226,
|
| 166 |
+
"eval_runtime": 550.439,
|
| 167 |
+
"eval_samples_per_second": 8.153,
|
| 168 |
+
"eval_steps_per_second": 2.038,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.049019607843137254,
|
| 173 |
+
"grad_norm": 13.235169410705566,
|
| 174 |
+
"learning_rate": 8e-05,
|
| 175 |
+
"loss": 0.2591,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.049019607843137254,
|
| 180 |
+
"eval_accuracy": 0.9035205245018005,
|
| 181 |
+
"eval_loss": 0.17823095619678497,
|
| 182 |
+
"eval_runtime": 543.3444,
|
| 183 |
+
"eval_samples_per_second": 8.26,
|
| 184 |
+
"eval_steps_per_second": 2.065,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.053475935828877004,
|
| 189 |
+
"grad_norm": 3.1415486335754395,
|
| 190 |
+
"learning_rate": 7.818181818181818e-05,
|
| 191 |
+
"loss": 0.261,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.053475935828877004,
|
| 196 |
+
"eval_accuracy": 0.8725489974021912,
|
| 197 |
+
"eval_loss": 0.23705999553203583,
|
| 198 |
+
"eval_runtime": 550.4034,
|
| 199 |
+
"eval_samples_per_second": 8.154,
|
| 200 |
+
"eval_steps_per_second": 2.039,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.057932263814616754,
|
| 205 |
+
"grad_norm": 0.12994514405727386,
|
| 206 |
+
"learning_rate": 7.636363636363637e-05,
|
| 207 |
+
"loss": 0.2626,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.057932263814616754,
|
| 212 |
+
"eval_accuracy": 0.89683598279953,
|
| 213 |
+
"eval_loss": 0.4889169931411743,
|
| 214 |
+
"eval_runtime": 545.9677,
|
| 215 |
+
"eval_samples_per_second": 8.22,
|
| 216 |
+
"eval_steps_per_second": 2.055,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.062388591800356503,
|
| 221 |
+
"grad_norm": 0.20874539017677307,
|
| 222 |
+
"learning_rate": 7.454545454545455e-05,
|
| 223 |
+
"loss": 0.3156,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.062388591800356503,
|
| 228 |
+
"eval_accuracy": 0.9021835923194885,
|
| 229 |
+
"eval_loss": 0.21060487627983093,
|
| 230 |
+
"eval_runtime": 543.8063,
|
| 231 |
+
"eval_samples_per_second": 8.253,
|
| 232 |
+
"eval_steps_per_second": 2.063,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06684491978609626,
|
| 237 |
+
"grad_norm": 1.1314120292663574,
|
| 238 |
+
"learning_rate": 7.272727272727273e-05,
|
| 239 |
+
"loss": 0.3342,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.06684491978609626,
|
| 244 |
+
"eval_accuracy": 0.9217914342880249,
|
| 245 |
+
"eval_loss": 0.19053570926189423,
|
| 246 |
+
"eval_runtime": 543.9184,
|
| 247 |
+
"eval_samples_per_second": 8.251,
|
| 248 |
+
"eval_steps_per_second": 2.063,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.07130124777183601,
|
| 253 |
+
"grad_norm": 5.4050092697143555,
|
| 254 |
+
"learning_rate": 7.090909090909092e-05,
|
| 255 |
+
"loss": 0.2658,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.07130124777183601,
|
| 260 |
+
"eval_accuracy": 0.9498662948608398,
|
| 261 |
+
"eval_loss": 0.12423694878816605,
|
| 262 |
+
"eval_runtime": 542.375,
|
| 263 |
+
"eval_samples_per_second": 8.275,
|
| 264 |
+
"eval_steps_per_second": 2.069,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.07575757575757576,
|
| 269 |
+
"grad_norm": 6.958705425262451,
|
| 270 |
+
"learning_rate": 6.90909090909091e-05,
|
| 271 |
+
"loss": 0.2162,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.07575757575757576,
|
| 276 |
+
"eval_accuracy": 0.9211229681968689,
|
| 277 |
+
"eval_loss": 0.1585531085729599,
|
| 278 |
+
"eval_runtime": 550.3192,
|
| 279 |
+
"eval_samples_per_second": 8.155,
|
| 280 |
+
"eval_steps_per_second": 2.039,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.08021390374331551,
|
| 285 |
+
"grad_norm": 6.347085475921631,
|
| 286 |
+
"learning_rate": 6.727272727272727e-05,
|
| 287 |
+
"loss": 0.2457,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.08021390374331551,
|
| 292 |
+
"eval_accuracy": 0.8422459959983826,
|
| 293 |
+
"eval_loss": 0.31698858737945557,
|
| 294 |
+
"eval_runtime": 550.4639,
|
| 295 |
+
"eval_samples_per_second": 8.153,
|
| 296 |
+
"eval_steps_per_second": 2.038,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.08467023172905526,
|
| 301 |
+
"grad_norm": 0.05648183450102806,
|
| 302 |
+
"learning_rate": 6.545454545454546e-05,
|
| 303 |
+
"loss": 0.178,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.08467023172905526,
|
| 308 |
+
"eval_accuracy": 0.9663547277450562,
|
| 309 |
+
"eval_loss": 0.08927226811647415,
|
| 310 |
+
"eval_runtime": 548.6704,
|
| 311 |
+
"eval_samples_per_second": 8.18,
|
| 312 |
+
"eval_steps_per_second": 2.045,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08912655971479501,
|
| 317 |
+
"grad_norm": 0.10795829445123672,
|
| 318 |
+
"learning_rate": 6.363636363636364e-05,
|
| 319 |
+
"loss": 0.2864,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.08912655971479501,
|
| 324 |
+
"eval_accuracy": 0.9079768061637878,
|
| 325 |
+
"eval_loss": 0.1990886628627777,
|
| 326 |
+
"eval_runtime": 547.327,
|
| 327 |
+
"eval_samples_per_second": 8.2,
|
| 328 |
+
"eval_steps_per_second": 2.05,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.09358288770053476,
|
| 333 |
+
"grad_norm": 0.10935252159833908,
|
| 334 |
+
"learning_rate": 6.181818181818182e-05,
|
| 335 |
+
"loss": 0.0852,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.09358288770053476,
|
| 340 |
+
"eval_accuracy": 0.9777183532714844,
|
| 341 |
+
"eval_loss": 0.06789236515760422,
|
| 342 |
+
"eval_runtime": 546.4384,
|
| 343 |
+
"eval_samples_per_second": 8.213,
|
| 344 |
+
"eval_steps_per_second": 2.053,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.09803921568627451,
|
| 349 |
+
"grad_norm": 0.724420428276062,
|
| 350 |
+
"learning_rate": 6e-05,
|
| 351 |
+
"loss": 0.2227,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.09803921568627451,
|
| 356 |
+
"eval_accuracy": 0.9676916003227234,
|
| 357 |
+
"eval_loss": 0.08267948776483536,
|
| 358 |
+
"eval_runtime": 543.5229,
|
| 359 |
+
"eval_samples_per_second": 8.257,
|
| 360 |
+
"eval_steps_per_second": 2.064,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.10249554367201426,
|
| 365 |
+
"grad_norm": 3.5009875297546387,
|
| 366 |
+
"learning_rate": 5.818181818181818e-05,
|
| 367 |
+
"loss": 0.0894,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.10249554367201426,
|
| 372 |
+
"eval_accuracy": 0.9799465537071228,
|
| 373 |
+
"eval_loss": 0.05659499019384384,
|
| 374 |
+
"eval_runtime": 547.0317,
|
| 375 |
+
"eval_samples_per_second": 8.204,
|
| 376 |
+
"eval_steps_per_second": 2.051,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.10695187165775401,
|
| 381 |
+
"grad_norm": 13.977231979370117,
|
| 382 |
+
"learning_rate": 5.636363636363636e-05,
|
| 383 |
+
"loss": 0.1766,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.10695187165775401,
|
| 388 |
+
"eval_accuracy": 0.9275846481323242,
|
| 389 |
+
"eval_loss": 0.17180828750133514,
|
| 390 |
+
"eval_runtime": 554.5064,
|
| 391 |
+
"eval_samples_per_second": 8.094,
|
| 392 |
+
"eval_steps_per_second": 2.023,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.11140819964349376,
|
| 397 |
+
"grad_norm": 1.5852386951446533,
|
| 398 |
+
"learning_rate": 5.4545454545454546e-05,
|
| 399 |
+
"loss": 0.1133,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.11140819964349376,
|
| 404 |
+
"eval_accuracy": 0.9625668525695801,
|
| 405 |
+
"eval_loss": 0.09958070516586304,
|
| 406 |
+
"eval_runtime": 546.2195,
|
| 407 |
+
"eval_samples_per_second": 8.216,
|
| 408 |
+
"eval_steps_per_second": 2.054,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.11586452762923351,
|
| 413 |
+
"grad_norm": 4.984052658081055,
|
| 414 |
+
"learning_rate": 5.272727272727272e-05,
|
| 415 |
+
"loss": 0.1581,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.11586452762923351,
|
| 420 |
+
"eval_accuracy": 0.9616755843162537,
|
| 421 |
+
"eval_loss": 0.09552862495183945,
|
| 422 |
+
"eval_runtime": 542.5097,
|
| 423 |
+
"eval_samples_per_second": 8.273,
|
| 424 |
+
"eval_steps_per_second": 2.068,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.12032085561497326,
|
| 429 |
+
"grad_norm": 0.06418484449386597,
|
| 430 |
+
"learning_rate": 5.090909090909091e-05,
|
| 431 |
+
"loss": 0.1164,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.12032085561497326,
|
| 436 |
+
"eval_accuracy": 0.9962121248245239,
|
| 437 |
+
"eval_loss": 0.017139658331871033,
|
| 438 |
+
"eval_runtime": 545.1464,
|
| 439 |
+
"eval_samples_per_second": 8.233,
|
| 440 |
+
"eval_steps_per_second": 2.058,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.12477718360071301,
|
| 445 |
+
"grad_norm": 1.2460925579071045,
|
| 446 |
+
"learning_rate": 4.909090909090909e-05,
|
| 447 |
+
"loss": 0.0199,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.12477718360071301,
|
| 452 |
+
"eval_accuracy": 0.9884135723114014,
|
| 453 |
+
"eval_loss": 0.04466737061738968,
|
| 454 |
+
"eval_runtime": 545.2406,
|
| 455 |
+
"eval_samples_per_second": 8.231,
|
| 456 |
+
"eval_steps_per_second": 2.058,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.12923351158645277,
|
| 461 |
+
"grad_norm": 0.021525170654058456,
|
| 462 |
+
"learning_rate": 4.7272727272727275e-05,
|
| 463 |
+
"loss": 0.0358,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.12923351158645277,
|
| 468 |
+
"eval_accuracy": 0.9625668525695801,
|
| 469 |
+
"eval_loss": 0.08889108896255493,
|
| 470 |
+
"eval_runtime": 551.6453,
|
| 471 |
+
"eval_samples_per_second": 8.136,
|
| 472 |
+
"eval_steps_per_second": 2.034,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.13368983957219252,
|
| 477 |
+
"grad_norm": 10.565823554992676,
|
| 478 |
+
"learning_rate": 4.545454545454546e-05,
|
| 479 |
+
"loss": 0.0134,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.13368983957219252,
|
| 484 |
+
"eval_accuracy": 0.9607843160629272,
|
| 485 |
+
"eval_loss": 0.11075662821531296,
|
| 486 |
+
"eval_runtime": 546.8222,
|
| 487 |
+
"eval_samples_per_second": 8.207,
|
| 488 |
+
"eval_steps_per_second": 2.052,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.13814616755793227,
|
| 493 |
+
"grad_norm": 2.769604444503784,
|
| 494 |
+
"learning_rate": 4.3636363636363636e-05,
|
| 495 |
+
"loss": 0.0085,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.13814616755793227,
|
| 500 |
+
"eval_accuracy": 0.9204545617103577,
|
| 501 |
+
"eval_loss": 0.24255433678627014,
|
| 502 |
+
"eval_runtime": 551.4065,
|
| 503 |
+
"eval_samples_per_second": 8.139,
|
| 504 |
+
"eval_steps_per_second": 2.035,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.14260249554367202,
|
| 509 |
+
"grad_norm": 0.027551617473363876,
|
| 510 |
+
"learning_rate": 4.181818181818182e-05,
|
| 511 |
+
"loss": 0.0691,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.14260249554367202,
|
| 516 |
+
"eval_accuracy": 0.9496434926986694,
|
| 517 |
+
"eval_loss": 0.15619446337223053,
|
| 518 |
+
"eval_runtime": 545.5513,
|
| 519 |
+
"eval_samples_per_second": 8.227,
|
| 520 |
+
"eval_steps_per_second": 2.057,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.14705882352941177,
|
| 525 |
+
"grad_norm": 0.010140771977603436,
|
| 526 |
+
"learning_rate": 4e-05,
|
| 527 |
+
"loss": 0.2242,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.14705882352941177,
|
| 532 |
+
"eval_accuracy": 0.9968805909156799,
|
| 533 |
+
"eval_loss": 0.01314464956521988,
|
| 534 |
+
"eval_runtime": 545.9537,
|
| 535 |
+
"eval_samples_per_second": 8.22,
|
| 536 |
+
"eval_steps_per_second": 2.055,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.15151515151515152,
|
| 541 |
+
"grad_norm": 0.011543634347617626,
|
| 542 |
+
"learning_rate": 3.818181818181819e-05,
|
| 543 |
+
"loss": 0.1593,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.15151515151515152,
|
| 548 |
+
"eval_accuracy": 0.9890819787979126,
|
| 549 |
+
"eval_loss": 0.03110310062766075,
|
| 550 |
+
"eval_runtime": 547.8599,
|
| 551 |
+
"eval_samples_per_second": 8.192,
|
| 552 |
+
"eval_steps_per_second": 2.048,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.15597147950089127,
|
| 557 |
+
"grad_norm": 0.06019105017185211,
|
| 558 |
+
"learning_rate": 3.6363636363636364e-05,
|
| 559 |
+
"loss": 0.0065,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.15597147950089127,
|
| 564 |
+
"eval_accuracy": 0.9643493890762329,
|
| 565 |
+
"eval_loss": 0.11227227747440338,
|
| 566 |
+
"eval_runtime": 546.9694,
|
| 567 |
+
"eval_samples_per_second": 8.205,
|
| 568 |
+
"eval_steps_per_second": 2.051,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.16042780748663102,
|
| 573 |
+
"grad_norm": 0.028353577479720116,
|
| 574 |
+
"learning_rate": 3.454545454545455e-05,
|
| 575 |
+
"loss": 0.0626,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.16042780748663102,
|
| 580 |
+
"eval_accuracy": 0.9817290306091309,
|
| 581 |
+
"eval_loss": 0.061965711414813995,
|
| 582 |
+
"eval_runtime": 552.3776,
|
| 583 |
+
"eval_samples_per_second": 8.125,
|
| 584 |
+
"eval_steps_per_second": 2.031,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.16488413547237077,
|
| 589 |
+
"grad_norm": 0.4727762043476105,
|
| 590 |
+
"learning_rate": 3.272727272727273e-05,
|
| 591 |
+
"loss": 0.0281,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.16488413547237077,
|
| 596 |
+
"eval_accuracy": 0.9596702456474304,
|
| 597 |
+
"eval_loss": 0.12898869812488556,
|
| 598 |
+
"eval_runtime": 546.7154,
|
| 599 |
+
"eval_samples_per_second": 8.209,
|
| 600 |
+
"eval_steps_per_second": 2.052,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.16934046345811052,
|
| 605 |
+
"grad_norm": 0.013356081210076809,
|
| 606 |
+
"learning_rate": 3.090909090909091e-05,
|
| 607 |
+
"loss": 0.0189,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.16934046345811052,
|
| 612 |
+
"eval_accuracy": 0.991310179233551,
|
| 613 |
+
"eval_loss": 0.02557438611984253,
|
| 614 |
+
"eval_runtime": 546.8803,
|
| 615 |
+
"eval_samples_per_second": 8.207,
|
| 616 |
+
"eval_steps_per_second": 2.052,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.17379679144385027,
|
| 621 |
+
"grad_norm": 0.056645121425390244,
|
| 622 |
+
"learning_rate": 2.909090909090909e-05,
|
| 623 |
+
"loss": 0.1307,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.17379679144385027,
|
| 628 |
+
"eval_accuracy": 0.9783868193626404,
|
| 629 |
+
"eval_loss": 0.07275046408176422,
|
| 630 |
+
"eval_runtime": 548.2067,
|
| 631 |
+
"eval_samples_per_second": 8.187,
|
| 632 |
+
"eval_steps_per_second": 2.047,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.17825311942959002,
|
| 637 |
+
"grad_norm": 0.0406530387699604,
|
| 638 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 639 |
+
"loss": 0.0061,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.17825311942959002,
|
| 644 |
+
"eval_accuracy": 0.9471924901008606,
|
| 645 |
+
"eval_loss": 0.17399340867996216,
|
| 646 |
+
"eval_runtime": 542.8312,
|
| 647 |
+
"eval_samples_per_second": 8.268,
|
| 648 |
+
"eval_steps_per_second": 2.067,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.18270944741532977,
|
| 653 |
+
"grad_norm": 0.07212503999471664,
|
| 654 |
+
"learning_rate": 2.5454545454545454e-05,
|
| 655 |
+
"loss": 0.0739,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.18270944741532977,
|
| 660 |
+
"eval_accuracy": 0.9500890970230103,
|
| 661 |
+
"eval_loss": 0.16757053136825562,
|
| 662 |
+
"eval_runtime": 542.9814,
|
| 663 |
+
"eval_samples_per_second": 8.265,
|
| 664 |
+
"eval_steps_per_second": 2.066,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.18716577540106952,
|
| 669 |
+
"grad_norm": 0.10422785580158234,
|
| 670 |
+
"learning_rate": 2.3636363636363637e-05,
|
| 671 |
+
"loss": 0.0028,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.18716577540106952,
|
| 676 |
+
"eval_accuracy": 0.9783868193626404,
|
| 677 |
+
"eval_loss": 0.07298260927200317,
|
| 678 |
+
"eval_runtime": 539.9794,
|
| 679 |
+
"eval_samples_per_second": 8.311,
|
| 680 |
+
"eval_steps_per_second": 2.078,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.19162210338680927,
|
| 685 |
+
"grad_norm": 0.007547458633780479,
|
| 686 |
+
"learning_rate": 2.1818181818181818e-05,
|
| 687 |
+
"loss": 0.0011,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.19162210338680927,
|
| 692 |
+
"eval_accuracy": 0.977495551109314,
|
| 693 |
+
"eval_loss": 0.07658497989177704,
|
| 694 |
+
"eval_runtime": 549.3337,
|
| 695 |
+
"eval_samples_per_second": 8.17,
|
| 696 |
+
"eval_steps_per_second": 2.042,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.19607843137254902,
|
| 701 |
+
"grad_norm": 0.007210019510239363,
|
| 702 |
+
"learning_rate": 2e-05,
|
| 703 |
+
"loss": 0.0019,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.19607843137254902,
|
| 708 |
+
"eval_accuracy": 0.9636809229850769,
|
| 709 |
+
"eval_loss": 0.11826927214860916,
|
| 710 |
+
"eval_runtime": 543.2619,
|
| 711 |
+
"eval_samples_per_second": 8.261,
|
| 712 |
+
"eval_steps_per_second": 2.065,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.20053475935828877,
|
| 717 |
+
"grad_norm": 0.013209226541221142,
|
| 718 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 719 |
+
"loss": 0.0388,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.20053475935828877,
|
| 724 |
+
"eval_accuracy": 0.9576649069786072,
|
| 725 |
+
"eval_loss": 0.14545659720897675,
|
| 726 |
+
"eval_runtime": 544.4489,
|
| 727 |
+
"eval_samples_per_second": 8.243,
|
| 728 |
+
"eval_steps_per_second": 2.061,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.20499108734402852,
|
| 733 |
+
"grad_norm": 0.007217989303171635,
|
| 734 |
+
"learning_rate": 1.6363636363636366e-05,
|
| 735 |
+
"loss": 0.0041,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.20499108734402852,
|
| 740 |
+
"eval_accuracy": 0.9523172974586487,
|
| 741 |
+
"eval_loss": 0.16972462832927704,
|
| 742 |
+
"eval_runtime": 545.2894,
|
| 743 |
+
"eval_samples_per_second": 8.23,
|
| 744 |
+
"eval_steps_per_second": 2.058,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.20944741532976827,
|
| 749 |
+
"grad_norm": 0.8174325227737427,
|
| 750 |
+
"learning_rate": 1.4545454545454545e-05,
|
| 751 |
+
"loss": 0.0064,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.20944741532976827,
|
| 756 |
+
"eval_accuracy": 0.9509803652763367,
|
| 757 |
+
"eval_loss": 0.17997007071971893,
|
| 758 |
+
"eval_runtime": 543.7245,
|
| 759 |
+
"eval_samples_per_second": 8.254,
|
| 760 |
+
"eval_steps_per_second": 2.064,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.21390374331550802,
|
| 765 |
+
"grad_norm": 0.032331835478544235,
|
| 766 |
+
"learning_rate": 1.2727272727272727e-05,
|
| 767 |
+
"loss": 0.0008,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.21390374331550802,
|
| 772 |
+
"eval_accuracy": 0.9610071182250977,
|
| 773 |
+
"eval_loss": 0.1397457718849182,
|
| 774 |
+
"eval_runtime": 546.6811,
|
| 775 |
+
"eval_samples_per_second": 8.21,
|
| 776 |
+
"eval_steps_per_second": 2.052,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.21836007130124777,
|
| 781 |
+
"grad_norm": 6.01271915435791,
|
| 782 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 783 |
+
"loss": 0.0688,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.21836007130124777,
|
| 788 |
+
"eval_accuracy": 0.9694741368293762,
|
| 789 |
+
"eval_loss": 0.10627125203609467,
|
| 790 |
+
"eval_runtime": 544.0334,
|
| 791 |
+
"eval_samples_per_second": 8.249,
|
| 792 |
+
"eval_steps_per_second": 2.062,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.22281639928698752,
|
| 797 |
+
"grad_norm": 0.004174210596829653,
|
| 798 |
+
"learning_rate": 9.090909090909091e-06,
|
| 799 |
+
"loss": 0.0008,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.22281639928698752,
|
| 804 |
+
"eval_accuracy": 0.9844028353691101,
|
| 805 |
+
"eval_loss": 0.05713631212711334,
|
| 806 |
+
"eval_runtime": 551.9061,
|
| 807 |
+
"eval_samples_per_second": 8.132,
|
| 808 |
+
"eval_steps_per_second": 2.033,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.22727272727272727,
|
| 813 |
+
"grad_norm": 0.006048465613275766,
|
| 814 |
+
"learning_rate": 7.272727272727272e-06,
|
| 815 |
+
"loss": 0.0018,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.22727272727272727,
|
| 820 |
+
"eval_accuracy": 0.9870766401290894,
|
| 821 |
+
"eval_loss": 0.048317644745111465,
|
| 822 |
+
"eval_runtime": 541.5344,
|
| 823 |
+
"eval_samples_per_second": 8.288,
|
| 824 |
+
"eval_steps_per_second": 2.072,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.23172905525846701,
|
| 829 |
+
"grad_norm": 13.882183074951172,
|
| 830 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 831 |
+
"loss": 0.1531,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.23172905525846701,
|
| 836 |
+
"eval_accuracy": 0.9870766401290894,
|
| 837 |
+
"eval_loss": 0.04793470725417137,
|
| 838 |
+
"eval_runtime": 540.1145,
|
| 839 |
+
"eval_samples_per_second": 8.309,
|
| 840 |
+
"eval_steps_per_second": 2.077,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.23618538324420676,
|
| 845 |
+
"grad_norm": 0.21645870804786682,
|
| 846 |
+
"learning_rate": 3.636363636363636e-06,
|
| 847 |
+
"loss": 0.0027,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.23618538324420676,
|
| 852 |
+
"eval_accuracy": 0.9848484992980957,
|
| 853 |
+
"eval_loss": 0.055766720324754715,
|
| 854 |
+
"eval_runtime": 545.2691,
|
| 855 |
+
"eval_samples_per_second": 8.231,
|
| 856 |
+
"eval_steps_per_second": 2.058,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.24064171122994651,
|
| 861 |
+
"grad_norm": 0.27186936140060425,
|
| 862 |
+
"learning_rate": 1.818181818181818e-06,
|
| 863 |
+
"loss": 0.001,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.24064171122994651,
|
| 868 |
+
"eval_accuracy": 0.9844028353691101,
|
| 869 |
+
"eval_loss": 0.059176359325647354,
|
| 870 |
+
"eval_runtime": 537.6305,
|
| 871 |
+
"eval_samples_per_second": 8.348,
|
| 872 |
+
"eval_steps_per_second": 2.087,
|
| 873 |
+
"step": 540
|
| 874 |
+
}
|
| 875 |
+
],
|
| 876 |
+
"logging_steps": 10,
|
| 877 |
+
"max_steps": 550,
|
| 878 |
+
"num_input_tokens_seen": 0,
|
| 879 |
+
"num_train_epochs": 1,
|
| 880 |
+
"save_steps": 10,
|
| 881 |
+
"stateful_callbacks": {
|
| 882 |
+
"TrainerControl": {
|
| 883 |
+
"args": {
|
| 884 |
+
"should_epoch_stop": false,
|
| 885 |
+
"should_evaluate": false,
|
| 886 |
+
"should_log": false,
|
| 887 |
+
"should_save": true,
|
| 888 |
+
"should_training_stop": false
|
| 889 |
+
},
|
| 890 |
+
"attributes": {}
|
| 891 |
+
}
|
| 892 |
+
},
|
| 893 |
+
"total_flos": 6.207949973692707e+17,
|
| 894 |
+
"train_batch_size": 4,
|
| 895 |
+
"trial_name": null,
|
| 896 |
+
"trial_params": null
|
| 897 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33e029d40d6a013eb05e095032ee0ff7c2ba463caceabd1a6913c0f6f5fac184
|
| 3 |
+
size 5112
|