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 +913 -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": "4.42.0.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 sha256:13e6fb81d259471ca808cced310d701f6a4baddc853aa6831356c313c3534646
|
| 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": 2.3023250102996826,
|
| 14 |
+
"learning_rate": 0.0002945454545454545,
|
| 15 |
+
"loss": 0.6742,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004207868714496108,
|
| 20 |
+
"eval_accuracy": 0.8339995741844177,
|
| 21 |
+
"eval_loss": 0.5872370004653931,
|
| 22 |
+
"eval_runtime": 559.2325,
|
| 23 |
+
"eval_samples_per_second": 8.499,
|
| 24 |
+
"eval_steps_per_second": 2.126,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.008415737428992216,
|
| 29 |
+
"grad_norm": 10.144115447998047,
|
| 30 |
+
"learning_rate": 0.00028909090909090904,
|
| 31 |
+
"loss": 0.7741,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.008415737428992216,
|
| 36 |
+
"eval_accuracy": 0.5914159417152405,
|
| 37 |
+
"eval_loss": 1.0286434888839722,
|
| 38 |
+
"eval_runtime": 548.7332,
|
| 39 |
+
"eval_samples_per_second": 8.662,
|
| 40 |
+
"eval_steps_per_second": 2.167,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.012623606143488323,
|
| 45 |
+
"grad_norm": 4.225183486938477,
|
| 46 |
+
"learning_rate": 0.0002836363636363636,
|
| 47 |
+
"loss": 0.8739,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.012623606143488323,
|
| 52 |
+
"eval_accuracy": 0.5914159417152405,
|
| 53 |
+
"eval_loss": 0.6784296631813049,
|
| 54 |
+
"eval_runtime": 552.4092,
|
| 55 |
+
"eval_samples_per_second": 8.604,
|
| 56 |
+
"eval_steps_per_second": 2.152,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.016831474857984433,
|
| 61 |
+
"grad_norm": 2.9214963912963867,
|
| 62 |
+
"learning_rate": 0.00027818181818181815,
|
| 63 |
+
"loss": 0.7206,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.016831474857984433,
|
| 68 |
+
"eval_accuracy": 0.5914159417152405,
|
| 69 |
+
"eval_loss": 0.6782186627388,
|
| 70 |
+
"eval_runtime": 553.9264,
|
| 71 |
+
"eval_samples_per_second": 8.581,
|
| 72 |
+
"eval_steps_per_second": 2.146,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.021039343572480537,
|
| 77 |
+
"grad_norm": 4.568026542663574,
|
| 78 |
+
"learning_rate": 0.0002727272727272727,
|
| 79 |
+
"loss": 0.7165,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.021039343572480537,
|
| 84 |
+
"eval_accuracy": 0.8114874958992004,
|
| 85 |
+
"eval_loss": 0.5903932452201843,
|
| 86 |
+
"eval_runtime": 556.1381,
|
| 87 |
+
"eval_samples_per_second": 8.546,
|
| 88 |
+
"eval_steps_per_second": 2.138,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.025247212286976645,
|
| 93 |
+
"grad_norm": 1.1793783903121948,
|
| 94 |
+
"learning_rate": 0.0002672727272727272,
|
| 95 |
+
"loss": 0.6391,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.025247212286976645,
|
| 100 |
+
"eval_accuracy": 0.5914159417152405,
|
| 101 |
+
"eval_loss": 0.6900256276130676,
|
| 102 |
+
"eval_runtime": 557.0755,
|
| 103 |
+
"eval_samples_per_second": 8.532,
|
| 104 |
+
"eval_steps_per_second": 2.134,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.029455081001472753,
|
| 109 |
+
"grad_norm": 1.104693055152893,
|
| 110 |
+
"learning_rate": 0.0002618181818181818,
|
| 111 |
+
"loss": 0.6694,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.029455081001472753,
|
| 116 |
+
"eval_accuracy": 0.8569324612617493,
|
| 117 |
+
"eval_loss": 0.6000075340270996,
|
| 118 |
+
"eval_runtime": 559.2951,
|
| 119 |
+
"eval_samples_per_second": 8.498,
|
| 120 |
+
"eval_steps_per_second": 2.126,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.033662949715968865,
|
| 125 |
+
"grad_norm": 3.541372537612915,
|
| 126 |
+
"learning_rate": 0.00025636363636363633,
|
| 127 |
+
"loss": 0.5766,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.033662949715968865,
|
| 132 |
+
"eval_accuracy": 0.8055964708328247,
|
| 133 |
+
"eval_loss": 0.5426038503646851,
|
| 134 |
+
"eval_runtime": 555.5844,
|
| 135 |
+
"eval_samples_per_second": 8.555,
|
| 136 |
+
"eval_steps_per_second": 2.14,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.03787081843046497,
|
| 141 |
+
"grad_norm": 3.705471992492676,
|
| 142 |
+
"learning_rate": 0.00025090909090909086,
|
| 143 |
+
"loss": 0.4908,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.03787081843046497,
|
| 148 |
+
"eval_accuracy": 0.8270565867424011,
|
| 149 |
+
"eval_loss": 0.39803212881088257,
|
| 150 |
+
"eval_runtime": 553.6423,
|
| 151 |
+
"eval_samples_per_second": 8.585,
|
| 152 |
+
"eval_steps_per_second": 2.148,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.042078687144961074,
|
| 157 |
+
"grad_norm": 2.6912827491760254,
|
| 158 |
+
"learning_rate": 0.00024545454545454545,
|
| 159 |
+
"loss": 0.5775,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.042078687144961074,
|
| 164 |
+
"eval_accuracy": 0.663791298866272,
|
| 165 |
+
"eval_loss": 0.7920271158218384,
|
| 166 |
+
"eval_runtime": 547.7028,
|
| 167 |
+
"eval_samples_per_second": 8.678,
|
| 168 |
+
"eval_steps_per_second": 2.171,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.046286555859457186,
|
| 173 |
+
"grad_norm": 1.2508213520050049,
|
| 174 |
+
"learning_rate": 0.00023999999999999998,
|
| 175 |
+
"loss": 0.5908,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.046286555859457186,
|
| 180 |
+
"eval_accuracy": 0.861561119556427,
|
| 181 |
+
"eval_loss": 0.3321113884449005,
|
| 182 |
+
"eval_runtime": 548.2222,
|
| 183 |
+
"eval_samples_per_second": 8.67,
|
| 184 |
+
"eval_steps_per_second": 2.169,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.05049442457395329,
|
| 189 |
+
"grad_norm": 3.2217397689819336,
|
| 190 |
+
"learning_rate": 0.00023454545454545454,
|
| 191 |
+
"loss": 0.3658,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.05049442457395329,
|
| 196 |
+
"eval_accuracy": 0.8897538185119629,
|
| 197 |
+
"eval_loss": 0.26357656717300415,
|
| 198 |
+
"eval_runtime": 553.9916,
|
| 199 |
+
"eval_samples_per_second": 8.58,
|
| 200 |
+
"eval_steps_per_second": 2.146,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.0547022932884494,
|
| 205 |
+
"grad_norm": 1.183059811592102,
|
| 206 |
+
"learning_rate": 0.00022909090909090907,
|
| 207 |
+
"loss": 0.4267,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.0547022932884494,
|
| 212 |
+
"eval_accuracy": 0.8424153327941895,
|
| 213 |
+
"eval_loss": 0.27429431676864624,
|
| 214 |
+
"eval_runtime": 552.9034,
|
| 215 |
+
"eval_samples_per_second": 8.596,
|
| 216 |
+
"eval_steps_per_second": 2.15,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.05891016200294551,
|
| 221 |
+
"grad_norm": 3.4732155799865723,
|
| 222 |
+
"learning_rate": 0.00022363636363636363,
|
| 223 |
+
"loss": 0.2191,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.05891016200294551,
|
| 228 |
+
"eval_accuracy": 0.9164738059043884,
|
| 229 |
+
"eval_loss": 0.28884753584861755,
|
| 230 |
+
"eval_runtime": 552.6166,
|
| 231 |
+
"eval_samples_per_second": 8.601,
|
| 232 |
+
"eval_steps_per_second": 2.152,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.06311803071744161,
|
| 237 |
+
"grad_norm": 2.1488490104675293,
|
| 238 |
+
"learning_rate": 0.00021818181818181816,
|
| 239 |
+
"loss": 0.25,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.06311803071744161,
|
| 244 |
+
"eval_accuracy": 0.9124763011932373,
|
| 245 |
+
"eval_loss": 0.36515381932258606,
|
| 246 |
+
"eval_runtime": 551.2736,
|
| 247 |
+
"eval_samples_per_second": 8.622,
|
| 248 |
+
"eval_steps_per_second": 2.157,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.06732589943193773,
|
| 253 |
+
"grad_norm": 2.5727055072784424,
|
| 254 |
+
"learning_rate": 0.00021272727272727272,
|
| 255 |
+
"loss": 0.4994,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.06732589943193773,
|
| 260 |
+
"eval_accuracy": 0.908478856086731,
|
| 261 |
+
"eval_loss": 0.24686028063297272,
|
| 262 |
+
"eval_runtime": 555.0552,
|
| 263 |
+
"eval_samples_per_second": 8.563,
|
| 264 |
+
"eval_steps_per_second": 2.142,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.07153376814643383,
|
| 269 |
+
"grad_norm": 2.7330338954925537,
|
| 270 |
+
"learning_rate": 0.00020727272727272725,
|
| 271 |
+
"loss": 0.274,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.07153376814643383,
|
| 276 |
+
"eval_accuracy": 0.9383547306060791,
|
| 277 |
+
"eval_loss": 0.16697467863559723,
|
| 278 |
+
"eval_runtime": 551.1246,
|
| 279 |
+
"eval_samples_per_second": 8.624,
|
| 280 |
+
"eval_steps_per_second": 2.157,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.07574163686092994,
|
| 285 |
+
"grad_norm": 21.431997299194336,
|
| 286 |
+
"learning_rate": 0.0002018181818181818,
|
| 287 |
+
"loss": 0.2721,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.07574163686092994,
|
| 292 |
+
"eval_accuracy": 0.9185777306556702,
|
| 293 |
+
"eval_loss": 0.2895517349243164,
|
| 294 |
+
"eval_runtime": 550.1769,
|
| 295 |
+
"eval_samples_per_second": 8.639,
|
| 296 |
+
"eval_steps_per_second": 2.161,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.07994950557542604,
|
| 301 |
+
"grad_norm": 6.493747711181641,
|
| 302 |
+
"learning_rate": 0.00019636363636363634,
|
| 303 |
+
"loss": 0.4545,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.07994950557542604,
|
| 308 |
+
"eval_accuracy": 0.9242583513259888,
|
| 309 |
+
"eval_loss": 0.32155150175094604,
|
| 310 |
+
"eval_runtime": 551.3203,
|
| 311 |
+
"eval_samples_per_second": 8.621,
|
| 312 |
+
"eval_steps_per_second": 2.157,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.08415737428992215,
|
| 317 |
+
"grad_norm": 3.465214490890503,
|
| 318 |
+
"learning_rate": 0.0001909090909090909,
|
| 319 |
+
"loss": 0.3221,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.08415737428992215,
|
| 324 |
+
"eval_accuracy": 0.9358299970626831,
|
| 325 |
+
"eval_loss": 0.20135502517223358,
|
| 326 |
+
"eval_runtime": 551.1975,
|
| 327 |
+
"eval_samples_per_second": 8.623,
|
| 328 |
+
"eval_steps_per_second": 2.157,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.08836524300441827,
|
| 333 |
+
"grad_norm": 4.727893352508545,
|
| 334 |
+
"learning_rate": 0.00018545454545454543,
|
| 335 |
+
"loss": 0.2348,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.08836524300441827,
|
| 340 |
+
"eval_accuracy": 0.9309909343719482,
|
| 341 |
+
"eval_loss": 0.2655259370803833,
|
| 342 |
+
"eval_runtime": 546.4467,
|
| 343 |
+
"eval_samples_per_second": 8.698,
|
| 344 |
+
"eval_steps_per_second": 2.176,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.09257311171891437,
|
| 349 |
+
"grad_norm": 2.3263726234436035,
|
| 350 |
+
"learning_rate": 0.00017999999999999998,
|
| 351 |
+
"loss": 0.3247,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.09257311171891437,
|
| 356 |
+
"eval_accuracy": 0.9417210221290588,
|
| 357 |
+
"eval_loss": 0.1690194457769394,
|
| 358 |
+
"eval_runtime": 546.9859,
|
| 359 |
+
"eval_samples_per_second": 8.689,
|
| 360 |
+
"eval_steps_per_second": 2.174,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.09678098043341048,
|
| 365 |
+
"grad_norm": 0.32978275418281555,
|
| 366 |
+
"learning_rate": 0.00017454545454545452,
|
| 367 |
+
"loss": 0.1945,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.09678098043341048,
|
| 372 |
+
"eval_accuracy": 0.897538423538208,
|
| 373 |
+
"eval_loss": 0.23890452086925507,
|
| 374 |
+
"eval_runtime": 556.5512,
|
| 375 |
+
"eval_samples_per_second": 8.54,
|
| 376 |
+
"eval_steps_per_second": 2.136,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.10098884914790658,
|
| 381 |
+
"grad_norm": 0.2658408582210541,
|
| 382 |
+
"learning_rate": 0.00016909090909090907,
|
| 383 |
+
"loss": 0.2405,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.10098884914790658,
|
| 388 |
+
"eval_accuracy": 0.9516094923019409,
|
| 389 |
+
"eval_loss": 0.1562691628932953,
|
| 390 |
+
"eval_runtime": 550.3268,
|
| 391 |
+
"eval_samples_per_second": 8.637,
|
| 392 |
+
"eval_steps_per_second": 2.161,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.1051967178624027,
|
| 397 |
+
"grad_norm": 2.5755348205566406,
|
| 398 |
+
"learning_rate": 0.0001636363636363636,
|
| 399 |
+
"loss": 0.1715,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.1051967178624027,
|
| 404 |
+
"eval_accuracy": 0.9362508058547974,
|
| 405 |
+
"eval_loss": 0.2467575967311859,
|
| 406 |
+
"eval_runtime": 552.8523,
|
| 407 |
+
"eval_samples_per_second": 8.597,
|
| 408 |
+
"eval_steps_per_second": 2.151,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1094045865768988,
|
| 413 |
+
"grad_norm": 4.770780563354492,
|
| 414 |
+
"learning_rate": 0.00015818181818181816,
|
| 415 |
+
"loss": 0.1866,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1094045865768988,
|
| 420 |
+
"eval_accuracy": 0.9654954671859741,
|
| 421 |
+
"eval_loss": 0.16653123497962952,
|
| 422 |
+
"eval_runtime": 553.6754,
|
| 423 |
+
"eval_samples_per_second": 8.584,
|
| 424 |
+
"eval_steps_per_second": 2.147,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.11361245529139491,
|
| 429 |
+
"grad_norm": 0.04907585307955742,
|
| 430 |
+
"learning_rate": 0.0001527272727272727,
|
| 431 |
+
"loss": 0.2001,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.11361245529139491,
|
| 436 |
+
"eval_accuracy": 0.9648643136024475,
|
| 437 |
+
"eval_loss": 0.168188214302063,
|
| 438 |
+
"eval_runtime": 558.4706,
|
| 439 |
+
"eval_samples_per_second": 8.511,
|
| 440 |
+
"eval_steps_per_second": 2.129,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.11782032400589101,
|
| 445 |
+
"grad_norm": 0.042652346193790436,
|
| 446 |
+
"learning_rate": 0.00014727272727272725,
|
| 447 |
+
"loss": 0.1359,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.11782032400589101,
|
| 452 |
+
"eval_accuracy": 0.9686513543128967,
|
| 453 |
+
"eval_loss": 0.1295313686132431,
|
| 454 |
+
"eval_runtime": 553.3401,
|
| 455 |
+
"eval_samples_per_second": 8.59,
|
| 456 |
+
"eval_steps_per_second": 2.149,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.12202819272038712,
|
| 461 |
+
"grad_norm": 0.10347462445497513,
|
| 462 |
+
"learning_rate": 0.0001418181818181818,
|
| 463 |
+
"loss": 0.1633,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.12202819272038712,
|
| 468 |
+
"eval_accuracy": 0.969072163105011,
|
| 469 |
+
"eval_loss": 0.12678129971027374,
|
| 470 |
+
"eval_runtime": 553.2115,
|
| 471 |
+
"eval_samples_per_second": 8.592,
|
| 472 |
+
"eval_steps_per_second": 2.149,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.12623606143488322,
|
| 477 |
+
"grad_norm": 0.07808654755353928,
|
| 478 |
+
"learning_rate": 0.00013636363636363634,
|
| 479 |
+
"loss": 0.0169,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.12623606143488322,
|
| 484 |
+
"eval_accuracy": 0.9709656834602356,
|
| 485 |
+
"eval_loss": 0.1480005532503128,
|
| 486 |
+
"eval_runtime": 556.5578,
|
| 487 |
+
"eval_samples_per_second": 8.54,
|
| 488 |
+
"eval_steps_per_second": 2.136,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.13044393014937933,
|
| 493 |
+
"grad_norm": 5.326434135437012,
|
| 494 |
+
"learning_rate": 0.0001309090909090909,
|
| 495 |
+
"loss": 0.2247,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.13044393014937933,
|
| 500 |
+
"eval_accuracy": 0.9625499844551086,
|
| 501 |
+
"eval_loss": 0.1905258446931839,
|
| 502 |
+
"eval_runtime": 561.2973,
|
| 503 |
+
"eval_samples_per_second": 8.468,
|
| 504 |
+
"eval_steps_per_second": 2.118,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.13465179886387546,
|
| 509 |
+
"grad_norm": 0.11235077679157257,
|
| 510 |
+
"learning_rate": 0.00012545454545454543,
|
| 511 |
+
"loss": 0.0609,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.13465179886387546,
|
| 516 |
+
"eval_accuracy": 0.9627603888511658,
|
| 517 |
+
"eval_loss": 0.18458549678325653,
|
| 518 |
+
"eval_runtime": 553.4853,
|
| 519 |
+
"eval_samples_per_second": 8.587,
|
| 520 |
+
"eval_steps_per_second": 2.148,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.13885966757837157,
|
| 525 |
+
"grad_norm": 0.11561981588602066,
|
| 526 |
+
"learning_rate": 0.00011999999999999999,
|
| 527 |
+
"loss": 0.0687,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.13885966757837157,
|
| 532 |
+
"eval_accuracy": 0.9713864922523499,
|
| 533 |
+
"eval_loss": 0.1371518075466156,
|
| 534 |
+
"eval_runtime": 546.6765,
|
| 535 |
+
"eval_samples_per_second": 8.694,
|
| 536 |
+
"eval_steps_per_second": 2.175,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.14306753629286767,
|
| 541 |
+
"grad_norm": 0.08527376502752304,
|
| 542 |
+
"learning_rate": 0.00011454545454545453,
|
| 543 |
+
"loss": 0.2002,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.14306753629286767,
|
| 548 |
+
"eval_accuracy": 0.9711760878562927,
|
| 549 |
+
"eval_loss": 0.13917790353298187,
|
| 550 |
+
"eval_runtime": 557.0042,
|
| 551 |
+
"eval_samples_per_second": 8.533,
|
| 552 |
+
"eval_steps_per_second": 2.135,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.14727540500736377,
|
| 557 |
+
"grad_norm": 0.09233067184686661,
|
| 558 |
+
"learning_rate": 0.00010909090909090908,
|
| 559 |
+
"loss": 0.0674,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.14727540500736377,
|
| 564 |
+
"eval_accuracy": 0.9596044421195984,
|
| 565 |
+
"eval_loss": 0.2038307785987854,
|
| 566 |
+
"eval_runtime": 551.7111,
|
| 567 |
+
"eval_samples_per_second": 8.615,
|
| 568 |
+
"eval_steps_per_second": 2.155,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.15148327372185988,
|
| 573 |
+
"grad_norm": 1.7652984857559204,
|
| 574 |
+
"learning_rate": 0.00010363636363636362,
|
| 575 |
+
"loss": 0.2449,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.15148327372185988,
|
| 580 |
+
"eval_accuracy": 0.9699137210845947,
|
| 581 |
+
"eval_loss": 0.1559370756149292,
|
| 582 |
+
"eval_runtime": 551.9133,
|
| 583 |
+
"eval_samples_per_second": 8.612,
|
| 584 |
+
"eval_steps_per_second": 2.154,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.15569114243635598,
|
| 589 |
+
"grad_norm": 0.8582727313041687,
|
| 590 |
+
"learning_rate": 9.818181818181817e-05,
|
| 591 |
+
"loss": 0.0794,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.15569114243635598,
|
| 596 |
+
"eval_accuracy": 0.9697033166885376,
|
| 597 |
+
"eval_loss": 0.10738001018762589,
|
| 598 |
+
"eval_runtime": 550.561,
|
| 599 |
+
"eval_samples_per_second": 8.633,
|
| 600 |
+
"eval_steps_per_second": 2.16,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.1598990111508521,
|
| 605 |
+
"grad_norm": 0.20773591101169586,
|
| 606 |
+
"learning_rate": 9.272727272727271e-05,
|
| 607 |
+
"loss": 0.3155,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.1598990111508521,
|
| 612 |
+
"eval_accuracy": 0.9701241254806519,
|
| 613 |
+
"eval_loss": 0.1280895173549652,
|
| 614 |
+
"eval_runtime": 553.1498,
|
| 615 |
+
"eval_samples_per_second": 8.593,
|
| 616 |
+
"eval_steps_per_second": 2.15,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.1641068798653482,
|
| 621 |
+
"grad_norm": 0.19470787048339844,
|
| 622 |
+
"learning_rate": 8.727272727272726e-05,
|
| 623 |
+
"loss": 0.0265,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.1641068798653482,
|
| 628 |
+
"eval_accuracy": 0.9617083668708801,
|
| 629 |
+
"eval_loss": 0.172711580991745,
|
| 630 |
+
"eval_runtime": 555.6513,
|
| 631 |
+
"eval_samples_per_second": 8.554,
|
| 632 |
+
"eval_steps_per_second": 2.14,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.1683147485798443,
|
| 637 |
+
"grad_norm": 2.273648262023926,
|
| 638 |
+
"learning_rate": 8.18181818181818e-05,
|
| 639 |
+
"loss": 0.1631,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.1683147485798443,
|
| 644 |
+
"eval_accuracy": 0.9726488590240479,
|
| 645 |
+
"eval_loss": 0.11321134120225906,
|
| 646 |
+
"eval_runtime": 552.9935,
|
| 647 |
+
"eval_samples_per_second": 8.595,
|
| 648 |
+
"eval_steps_per_second": 2.15,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.17252261729434043,
|
| 653 |
+
"grad_norm": 0.14013071358203888,
|
| 654 |
+
"learning_rate": 7.636363636363635e-05,
|
| 655 |
+
"loss": 0.0432,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.17252261729434043,
|
| 660 |
+
"eval_accuracy": 0.9724384546279907,
|
| 661 |
+
"eval_loss": 0.0995568335056305,
|
| 662 |
+
"eval_runtime": 551.9628,
|
| 663 |
+
"eval_samples_per_second": 8.611,
|
| 664 |
+
"eval_steps_per_second": 2.154,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.17673048600883653,
|
| 669 |
+
"grad_norm": 5.5448994636535645,
|
| 670 |
+
"learning_rate": 7.09090909090909e-05,
|
| 671 |
+
"loss": 0.0878,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.17673048600883653,
|
| 676 |
+
"eval_accuracy": 0.9720176458358765,
|
| 677 |
+
"eval_loss": 0.11234048753976822,
|
| 678 |
+
"eval_runtime": 554.8087,
|
| 679 |
+
"eval_samples_per_second": 8.567,
|
| 680 |
+
"eval_steps_per_second": 2.143,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.18093835472333264,
|
| 685 |
+
"grad_norm": 0.11822725087404251,
|
| 686 |
+
"learning_rate": 6.545454545454545e-05,
|
| 687 |
+
"loss": 0.1447,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.18093835472333264,
|
| 692 |
+
"eval_accuracy": 0.9713864922523499,
|
| 693 |
+
"eval_loss": 0.10238009691238403,
|
| 694 |
+
"eval_runtime": 553.3152,
|
| 695 |
+
"eval_samples_per_second": 8.59,
|
| 696 |
+
"eval_steps_per_second": 2.149,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.18514622343782874,
|
| 701 |
+
"grad_norm": 0.1717545986175537,
|
| 702 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 703 |
+
"loss": 0.1208,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.18514622343782874,
|
| 708 |
+
"eval_accuracy": 0.971596896648407,
|
| 709 |
+
"eval_loss": 0.09492182731628418,
|
| 710 |
+
"eval_runtime": 550.8446,
|
| 711 |
+
"eval_samples_per_second": 8.629,
|
| 712 |
+
"eval_steps_per_second": 2.159,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.18935409215232485,
|
| 717 |
+
"grad_norm": 0.7529041171073914,
|
| 718 |
+
"learning_rate": 5.454545454545454e-05,
|
| 719 |
+
"loss": 0.0643,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.18935409215232485,
|
| 724 |
+
"eval_accuracy": 0.9720176458358765,
|
| 725 |
+
"eval_loss": 0.08474315702915192,
|
| 726 |
+
"eval_runtime": 558.3353,
|
| 727 |
+
"eval_samples_per_second": 8.513,
|
| 728 |
+
"eval_steps_per_second": 2.13,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.19356196086682095,
|
| 733 |
+
"grad_norm": 0.20409902930259705,
|
| 734 |
+
"learning_rate": 4.9090909090909084e-05,
|
| 735 |
+
"loss": 0.0433,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.19356196086682095,
|
| 740 |
+
"eval_accuracy": 0.9758047461509705,
|
| 741 |
+
"eval_loss": 0.0837019681930542,
|
| 742 |
+
"eval_runtime": 555.0163,
|
| 743 |
+
"eval_samples_per_second": 8.564,
|
| 744 |
+
"eval_steps_per_second": 2.142,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.19776982958131706,
|
| 749 |
+
"grad_norm": 0.052959345281124115,
|
| 750 |
+
"learning_rate": 4.363636363636363e-05,
|
| 751 |
+
"loss": 0.1001,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.19776982958131706,
|
| 756 |
+
"eval_accuracy": 0.9793814420700073,
|
| 757 |
+
"eval_loss": 0.08561240136623383,
|
| 758 |
+
"eval_runtime": 550.605,
|
| 759 |
+
"eval_samples_per_second": 8.632,
|
| 760 |
+
"eval_steps_per_second": 2.159,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.20197769829581316,
|
| 765 |
+
"grad_norm": 0.19414858520030975,
|
| 766 |
+
"learning_rate": 3.8181818181818174e-05,
|
| 767 |
+
"loss": 0.0887,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.20197769829581316,
|
| 772 |
+
"eval_accuracy": 0.9768567085266113,
|
| 773 |
+
"eval_loss": 0.11111757904291153,
|
| 774 |
+
"eval_runtime": 549.5968,
|
| 775 |
+
"eval_samples_per_second": 8.648,
|
| 776 |
+
"eval_steps_per_second": 2.163,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.20618556701030927,
|
| 781 |
+
"grad_norm": 0.03906171768903732,
|
| 782 |
+
"learning_rate": 3.2727272727272725e-05,
|
| 783 |
+
"loss": 0.0554,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.20618556701030927,
|
| 788 |
+
"eval_accuracy": 0.9758047461509705,
|
| 789 |
+
"eval_loss": 0.12151744961738586,
|
| 790 |
+
"eval_runtime": 552.2439,
|
| 791 |
+
"eval_samples_per_second": 8.607,
|
| 792 |
+
"eval_steps_per_second": 2.153,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2103934357248054,
|
| 797 |
+
"grad_norm": 0.08566837012767792,
|
| 798 |
+
"learning_rate": 2.727272727272727e-05,
|
| 799 |
+
"loss": 0.0293,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2103934357248054,
|
| 804 |
+
"eval_accuracy": 0.975383996963501,
|
| 805 |
+
"eval_loss": 0.12937308847904205,
|
| 806 |
+
"eval_runtime": 558.1401,
|
| 807 |
+
"eval_samples_per_second": 8.516,
|
| 808 |
+
"eval_steps_per_second": 2.13,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.2146013044393015,
|
| 813 |
+
"grad_norm": 0.09275167435407639,
|
| 814 |
+
"learning_rate": 2.1818181818181814e-05,
|
| 815 |
+
"loss": 0.0152,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.2146013044393015,
|
| 820 |
+
"eval_accuracy": 0.9758047461509705,
|
| 821 |
+
"eval_loss": 0.12702897191047668,
|
| 822 |
+
"eval_runtime": 555.0185,
|
| 823 |
+
"eval_samples_per_second": 8.564,
|
| 824 |
+
"eval_steps_per_second": 2.142,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.2188091731537976,
|
| 829 |
+
"grad_norm": 0.031154340133070946,
|
| 830 |
+
"learning_rate": 1.6363636363636363e-05,
|
| 831 |
+
"loss": 0.1286,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.2188091731537976,
|
| 836 |
+
"eval_accuracy": 0.9747527837753296,
|
| 837 |
+
"eval_loss": 0.133098304271698,
|
| 838 |
+
"eval_runtime": 554.216,
|
| 839 |
+
"eval_samples_per_second": 8.576,
|
| 840 |
+
"eval_steps_per_second": 2.145,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.22301704186829371,
|
| 845 |
+
"grad_norm": 0.031359899789094925,
|
| 846 |
+
"learning_rate": 1.0909090909090907e-05,
|
| 847 |
+
"loss": 0.0186,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.22301704186829371,
|
| 852 |
+
"eval_accuracy": 0.9737008213996887,
|
| 853 |
+
"eval_loss": 0.1388082504272461,
|
| 854 |
+
"eval_runtime": 549.5036,
|
| 855 |
+
"eval_samples_per_second": 8.65,
|
| 856 |
+
"eval_steps_per_second": 2.164,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.22722491058278982,
|
| 861 |
+
"grad_norm": 0.03772009164094925,
|
| 862 |
+
"learning_rate": 5.454545454545454e-06,
|
| 863 |
+
"loss": 0.0299,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.22722491058278982,
|
| 868 |
+
"eval_accuracy": 0.9732800126075745,
|
| 869 |
+
"eval_loss": 0.14506614208221436,
|
| 870 |
+
"eval_runtime": 554.9274,
|
| 871 |
+
"eval_samples_per_second": 8.565,
|
| 872 |
+
"eval_steps_per_second": 2.143,
|
| 873 |
+
"step": 540
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.23143277929728592,
|
| 877 |
+
"grad_norm": 0.07556959241628647,
|
| 878 |
+
"learning_rate": 0.0,
|
| 879 |
+
"loss": 0.0536,
|
| 880 |
+
"step": 550
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.23143277929728592,
|
| 884 |
+
"eval_accuracy": 0.972859263420105,
|
| 885 |
+
"eval_loss": 0.14720246195793152,
|
| 886 |
+
"eval_runtime": 548.0681,
|
| 887 |
+
"eval_samples_per_second": 8.672,
|
| 888 |
+
"eval_steps_per_second": 2.169,
|
| 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 sha256:48b6dd7c12f40ce8ed6bdbfb4b365f9b2e4a728bdd4abe11b2c5c357b12743a7
|
| 3 |
+
size 5112
|