Upload 9 files
Browse files- config.json +162 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1291 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,162 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-base",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForSpeechClassification"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 256,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": false,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "sum",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": false,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_norm": "group",
|
| 51 |
+
"feat_proj_dropout": 0.1,
|
| 52 |
+
"feat_quantizer_dropout": 0.0,
|
| 53 |
+
"final_dropout": 0.0,
|
| 54 |
+
"finetuning_task": "wav2vec2_clf",
|
| 55 |
+
"freeze_feat_extract_train": true,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_size": 768,
|
| 59 |
+
"id2label": {
|
| 60 |
+
"0": "animal",
|
| 61 |
+
"1": "bar-restaurant-nightclub",
|
| 62 |
+
"2": "boda-boda-motocyle",
|
| 63 |
+
"3": "car-alarm",
|
| 64 |
+
"4": "car-or-truck",
|
| 65 |
+
"5": "community-radio",
|
| 66 |
+
"6": "construction-site",
|
| 67 |
+
"7": "crowd-noise",
|
| 68 |
+
"8": "fabrication-workshop",
|
| 69 |
+
"9": "generator",
|
| 70 |
+
"10": "hawker-vendor",
|
| 71 |
+
"11": "herbalists",
|
| 72 |
+
"12": "mobile-music",
|
| 73 |
+
"13": "motorvehicle-horn",
|
| 74 |
+
"14": "motorvehicle-siren",
|
| 75 |
+
"15": "other",
|
| 76 |
+
"16": "regilious-venue",
|
| 77 |
+
"17": "school",
|
| 78 |
+
"18": "street-preacher"
|
| 79 |
+
},
|
| 80 |
+
"initializer_range": 0.02,
|
| 81 |
+
"intermediate_size": 3072,
|
| 82 |
+
"label2id": {
|
| 83 |
+
"animal": 0,
|
| 84 |
+
"bar-restaurant-nightclub": 1,
|
| 85 |
+
"boda-boda-motocyle": 2,
|
| 86 |
+
"car-alarm": 3,
|
| 87 |
+
"car-or-truck": 4,
|
| 88 |
+
"community-radio": 5,
|
| 89 |
+
"construction-site": 6,
|
| 90 |
+
"crowd-noise": 7,
|
| 91 |
+
"fabrication-workshop": 8,
|
| 92 |
+
"generator": 9,
|
| 93 |
+
"hawker-vendor": 10,
|
| 94 |
+
"herbalists": 11,
|
| 95 |
+
"mobile-music": 12,
|
| 96 |
+
"motorvehicle-horn": 13,
|
| 97 |
+
"motorvehicle-siren": 14,
|
| 98 |
+
"other": 15,
|
| 99 |
+
"regilious-venue": 16,
|
| 100 |
+
"school": 17,
|
| 101 |
+
"street-preacher": 18
|
| 102 |
+
},
|
| 103 |
+
"layer_norm_eps": 1e-05,
|
| 104 |
+
"layerdrop": 0.0,
|
| 105 |
+
"mask_channel_length": 10,
|
| 106 |
+
"mask_channel_min_space": 1,
|
| 107 |
+
"mask_channel_other": 0.0,
|
| 108 |
+
"mask_channel_prob": 0.0,
|
| 109 |
+
"mask_channel_selection": "static",
|
| 110 |
+
"mask_feature_length": 10,
|
| 111 |
+
"mask_feature_min_masks": 0,
|
| 112 |
+
"mask_feature_prob": 0.0,
|
| 113 |
+
"mask_time_length": 10,
|
| 114 |
+
"mask_time_min_masks": 2,
|
| 115 |
+
"mask_time_min_space": 1,
|
| 116 |
+
"mask_time_other": 0.0,
|
| 117 |
+
"mask_time_prob": 0.05,
|
| 118 |
+
"mask_time_selection": "static",
|
| 119 |
+
"model_type": "wav2vec2",
|
| 120 |
+
"no_mask_channel_overlap": false,
|
| 121 |
+
"no_mask_time_overlap": false,
|
| 122 |
+
"num_adapter_layers": 3,
|
| 123 |
+
"num_attention_heads": 12,
|
| 124 |
+
"num_codevector_groups": 2,
|
| 125 |
+
"num_codevectors_per_group": 320,
|
| 126 |
+
"num_conv_pos_embedding_groups": 16,
|
| 127 |
+
"num_conv_pos_embeddings": 128,
|
| 128 |
+
"num_feat_extract_layers": 7,
|
| 129 |
+
"num_hidden_layers": 12,
|
| 130 |
+
"num_negatives": 100,
|
| 131 |
+
"output_hidden_size": 768,
|
| 132 |
+
"pad_token_id": 0,
|
| 133 |
+
"pooling_mode": "mean",
|
| 134 |
+
"problem_type": "single_label_classification",
|
| 135 |
+
"proj_codevector_dim": 256,
|
| 136 |
+
"tdnn_dilation": [
|
| 137 |
+
1,
|
| 138 |
+
2,
|
| 139 |
+
3,
|
| 140 |
+
1,
|
| 141 |
+
1
|
| 142 |
+
],
|
| 143 |
+
"tdnn_dim": [
|
| 144 |
+
512,
|
| 145 |
+
512,
|
| 146 |
+
512,
|
| 147 |
+
512,
|
| 148 |
+
1500
|
| 149 |
+
],
|
| 150 |
+
"tdnn_kernel": [
|
| 151 |
+
5,
|
| 152 |
+
3,
|
| 153 |
+
3,
|
| 154 |
+
1,
|
| 155 |
+
1
|
| 156 |
+
],
|
| 157 |
+
"torch_dtype": "float32",
|
| 158 |
+
"transformers_version": "4.16.2",
|
| 159 |
+
"use_weighted_layer_sum": false,
|
| 160 |
+
"vocab_size": 32,
|
| 161 |
+
"xvector_output_dim": 512
|
| 162 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0988acfb8f8dff61386e95f97a82d5e481f8b116e3f15a2f0996e7b50118d41d
|
| 3 |
+
size 726332986
|
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": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e7e71dbcd8707c10e8a413b4c6603e29d7a3c2063b5f27b8ad04b3cf52366cf
|
| 3 |
+
size 379995494
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57906c5e7a31f75f86ee3c6ab67cafc214fe2020e2db95eaaf271c19c47df3b8
|
| 3 |
+
size 14308
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72660203d90f6cb60c1a1338710c71b4f397ad8361a2ca78f913c5fbbcd15871
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c981c29d2227bfb31336a0c27616a6213f5177915224c1c77b5134b4d439f1b9
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1291 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.4613642796490236,
|
| 5 |
+
"global_step": 42800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 2.4900000000000002e-05,
|
| 13 |
+
"loss": 1.6261,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"eval_accuracy": 0.4594048261642456,
|
| 19 |
+
"eval_loss": 1.6715384721755981,
|
| 20 |
+
"eval_runtime": 1195.9203,
|
| 21 |
+
"eval_samples_per_second": 10.34,
|
| 22 |
+
"eval_steps_per_second": 5.17,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08,
|
| 27 |
+
"learning_rate": 4.9850000000000006e-05,
|
| 28 |
+
"loss": 1.5983,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.08,
|
| 33 |
+
"eval_accuracy": 0.4273815155029297,
|
| 34 |
+
"eval_loss": 1.8006267547607422,
|
| 35 |
+
"eval_runtime": 1200.9941,
|
| 36 |
+
"eval_samples_per_second": 10.296,
|
| 37 |
+
"eval_steps_per_second": 5.148,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"learning_rate": 7.475000000000001e-05,
|
| 43 |
+
"loss": 1.8724,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12,
|
| 48 |
+
"eval_accuracy": 0.3767588436603546,
|
| 49 |
+
"eval_loss": 1.967054009437561,
|
| 50 |
+
"eval_runtime": 1198.349,
|
| 51 |
+
"eval_samples_per_second": 10.319,
|
| 52 |
+
"eval_steps_per_second": 5.16,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.16,
|
| 57 |
+
"learning_rate": 9.975000000000001e-05,
|
| 58 |
+
"loss": 1.9731,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16,
|
| 63 |
+
"eval_accuracy": 0.3982694447040558,
|
| 64 |
+
"eval_loss": 1.9361200332641602,
|
| 65 |
+
"eval_runtime": 1206.207,
|
| 66 |
+
"eval_samples_per_second": 10.252,
|
| 67 |
+
"eval_steps_per_second": 5.126,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.2,
|
| 72 |
+
"learning_rate": 9.979820627802691e-05,
|
| 73 |
+
"loss": 1.9923,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.2,
|
| 78 |
+
"eval_accuracy": 0.3914766311645508,
|
| 79 |
+
"eval_loss": 2.0836076736450195,
|
| 80 |
+
"eval_runtime": 1190.0672,
|
| 81 |
+
"eval_samples_per_second": 10.391,
|
| 82 |
+
"eval_steps_per_second": 5.196,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.24,
|
| 87 |
+
"learning_rate": 9.959437423562985e-05,
|
| 88 |
+
"loss": 1.8714,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.24,
|
| 93 |
+
"eval_accuracy": 0.47185832262039185,
|
| 94 |
+
"eval_loss": 1.6872302293777466,
|
| 95 |
+
"eval_runtime": 1192.2479,
|
| 96 |
+
"eval_samples_per_second": 10.372,
|
| 97 |
+
"eval_steps_per_second": 5.186,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.28,
|
| 102 |
+
"learning_rate": 9.939054219323278e-05,
|
| 103 |
+
"loss": 1.7592,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.28,
|
| 108 |
+
"eval_accuracy": 0.4551997482776642,
|
| 109 |
+
"eval_loss": 1.7559692859649658,
|
| 110 |
+
"eval_runtime": 1199.3651,
|
| 111 |
+
"eval_samples_per_second": 10.31,
|
| 112 |
+
"eval_steps_per_second": 5.155,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.32,
|
| 117 |
+
"learning_rate": 9.918671015083571e-05,
|
| 118 |
+
"loss": 1.6757,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.32,
|
| 123 |
+
"eval_accuracy": 0.5049328804016113,
|
| 124 |
+
"eval_loss": 1.6301194429397583,
|
| 125 |
+
"eval_runtime": 1201.6603,
|
| 126 |
+
"eval_samples_per_second": 10.291,
|
| 127 |
+
"eval_steps_per_second": 5.145,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.36,
|
| 132 |
+
"learning_rate": 9.898287810843865e-05,
|
| 133 |
+
"loss": 1.6694,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.36,
|
| 138 |
+
"eval_accuracy": 0.5016981959342957,
|
| 139 |
+
"eval_loss": 1.7610756158828735,
|
| 140 |
+
"eval_runtime": 1198.381,
|
| 141 |
+
"eval_samples_per_second": 10.319,
|
| 142 |
+
"eval_steps_per_second": 5.159,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.4,
|
| 147 |
+
"learning_rate": 9.877904606604159e-05,
|
| 148 |
+
"loss": 1.6181,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4,
|
| 153 |
+
"eval_accuracy": 0.5145560503005981,
|
| 154 |
+
"eval_loss": 1.5850120782852173,
|
| 155 |
+
"eval_runtime": 1202.0306,
|
| 156 |
+
"eval_samples_per_second": 10.288,
|
| 157 |
+
"eval_steps_per_second": 5.144,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.44,
|
| 162 |
+
"learning_rate": 9.857562168772932e-05,
|
| 163 |
+
"loss": 1.686,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.44,
|
| 168 |
+
"eval_accuracy": 0.5050137639045715,
|
| 169 |
+
"eval_loss": 1.6993136405944824,
|
| 170 |
+
"eval_runtime": 1206.9484,
|
| 171 |
+
"eval_samples_per_second": 10.246,
|
| 172 |
+
"eval_steps_per_second": 5.123,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.49,
|
| 177 |
+
"learning_rate": 9.837178964533225e-05,
|
| 178 |
+
"loss": 1.5614,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.49,
|
| 183 |
+
"eval_accuracy": 0.5293546915054321,
|
| 184 |
+
"eval_loss": 1.5649951696395874,
|
| 185 |
+
"eval_runtime": 1211.1061,
|
| 186 |
+
"eval_samples_per_second": 10.211,
|
| 187 |
+
"eval_steps_per_second": 5.105,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.53,
|
| 192 |
+
"learning_rate": 9.816795760293519e-05,
|
| 193 |
+
"loss": 1.5995,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.53,
|
| 198 |
+
"eval_accuracy": 0.5736697316169739,
|
| 199 |
+
"eval_loss": 1.4089938402175903,
|
| 200 |
+
"eval_runtime": 1208.3994,
|
| 201 |
+
"eval_samples_per_second": 10.233,
|
| 202 |
+
"eval_steps_per_second": 5.117,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.57,
|
| 207 |
+
"learning_rate": 9.796412556053812e-05,
|
| 208 |
+
"loss": 1.4745,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.57,
|
| 213 |
+
"eval_accuracy": 0.5655021667480469,
|
| 214 |
+
"eval_loss": 1.4745545387268066,
|
| 215 |
+
"eval_runtime": 1213.8281,
|
| 216 |
+
"eval_samples_per_second": 10.188,
|
| 217 |
+
"eval_steps_per_second": 5.094,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.61,
|
| 222 |
+
"learning_rate": 9.776029351814105e-05,
|
| 223 |
+
"loss": 1.4518,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.61,
|
| 228 |
+
"eval_accuracy": 0.5529677867889404,
|
| 229 |
+
"eval_loss": 1.5653101205825806,
|
| 230 |
+
"eval_runtime": 1212.4544,
|
| 231 |
+
"eval_samples_per_second": 10.199,
|
| 232 |
+
"eval_steps_per_second": 5.1,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.65,
|
| 237 |
+
"learning_rate": 9.755646147574398e-05,
|
| 238 |
+
"loss": 1.4693,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.65,
|
| 243 |
+
"eval_accuracy": 0.5558789968490601,
|
| 244 |
+
"eval_loss": 1.499045729637146,
|
| 245 |
+
"eval_runtime": 1219.4703,
|
| 246 |
+
"eval_samples_per_second": 10.14,
|
| 247 |
+
"eval_steps_per_second": 5.07,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.69,
|
| 252 |
+
"learning_rate": 9.735262943334693e-05,
|
| 253 |
+
"loss": 1.4723,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.69,
|
| 258 |
+
"eval_accuracy": 0.5532912611961365,
|
| 259 |
+
"eval_loss": 1.5437432527542114,
|
| 260 |
+
"eval_runtime": 1217.1159,
|
| 261 |
+
"eval_samples_per_second": 10.16,
|
| 262 |
+
"eval_steps_per_second": 5.08,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.73,
|
| 267 |
+
"learning_rate": 9.714879739094986e-05,
|
| 268 |
+
"loss": 1.468,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.73,
|
| 273 |
+
"eval_accuracy": 0.5289503335952759,
|
| 274 |
+
"eval_loss": 1.6594600677490234,
|
| 275 |
+
"eval_runtime": 1215.7239,
|
| 276 |
+
"eval_samples_per_second": 10.172,
|
| 277 |
+
"eval_steps_per_second": 5.086,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.77,
|
| 282 |
+
"learning_rate": 9.69449653485528e-05,
|
| 283 |
+
"loss": 1.4135,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.77,
|
| 288 |
+
"eval_accuracy": 0.5919456481933594,
|
| 289 |
+
"eval_loss": 1.2978076934814453,
|
| 290 |
+
"eval_runtime": 1220.3558,
|
| 291 |
+
"eval_samples_per_second": 10.133,
|
| 292 |
+
"eval_steps_per_second": 5.067,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.81,
|
| 297 |
+
"learning_rate": 9.674194863432532e-05,
|
| 298 |
+
"loss": 1.4042,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.81,
|
| 303 |
+
"eval_accuracy": 0.5769044160842896,
|
| 304 |
+
"eval_loss": 1.4572824239730835,
|
| 305 |
+
"eval_runtime": 1236.4044,
|
| 306 |
+
"eval_samples_per_second": 10.002,
|
| 307 |
+
"eval_steps_per_second": 5.001,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.85,
|
| 312 |
+
"learning_rate": 9.653811659192825e-05,
|
| 313 |
+
"loss": 1.3864,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.85,
|
| 318 |
+
"eval_accuracy": 0.5811903476715088,
|
| 319 |
+
"eval_loss": 1.4342533349990845,
|
| 320 |
+
"eval_runtime": 1219.3647,
|
| 321 |
+
"eval_samples_per_second": 10.141,
|
| 322 |
+
"eval_steps_per_second": 5.071,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.89,
|
| 327 |
+
"learning_rate": 9.63342845495312e-05,
|
| 328 |
+
"loss": 1.3847,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.89,
|
| 333 |
+
"eval_accuracy": 0.6037521958351135,
|
| 334 |
+
"eval_loss": 1.433942437171936,
|
| 335 |
+
"eval_runtime": 1217.0747,
|
| 336 |
+
"eval_samples_per_second": 10.16,
|
| 337 |
+
"eval_steps_per_second": 5.08,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.93,
|
| 342 |
+
"learning_rate": 9.613045250713413e-05,
|
| 343 |
+
"loss": 1.3346,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.93,
|
| 348 |
+
"eval_accuracy": 0.6101406812667847,
|
| 349 |
+
"eval_loss": 1.3803991079330444,
|
| 350 |
+
"eval_runtime": 1217.1935,
|
| 351 |
+
"eval_samples_per_second": 10.159,
|
| 352 |
+
"eval_steps_per_second": 5.08,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.97,
|
| 357 |
+
"learning_rate": 9.592662046473706e-05,
|
| 358 |
+
"loss": 1.3487,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.97,
|
| 363 |
+
"eval_accuracy": 0.6148309707641602,
|
| 364 |
+
"eval_loss": 1.342442274093628,
|
| 365 |
+
"eval_runtime": 1216.3216,
|
| 366 |
+
"eval_samples_per_second": 10.167,
|
| 367 |
+
"eval_steps_per_second": 5.083,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.01,
|
| 372 |
+
"learning_rate": 9.572278842233999e-05,
|
| 373 |
+
"loss": 1.3597,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.01,
|
| 378 |
+
"eval_accuracy": 0.6078764200210571,
|
| 379 |
+
"eval_loss": 1.3775306940078735,
|
| 380 |
+
"eval_runtime": 1216.7531,
|
| 381 |
+
"eval_samples_per_second": 10.163,
|
| 382 |
+
"eval_steps_per_second": 5.082,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.05,
|
| 387 |
+
"learning_rate": 9.551895637994293e-05,
|
| 388 |
+
"loss": 1.3021,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.05,
|
| 393 |
+
"eval_accuracy": 0.6058547496795654,
|
| 394 |
+
"eval_loss": 1.3221111297607422,
|
| 395 |
+
"eval_runtime": 1220.1747,
|
| 396 |
+
"eval_samples_per_second": 10.135,
|
| 397 |
+
"eval_steps_per_second": 5.067,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.09,
|
| 402 |
+
"learning_rate": 9.531512433754587e-05,
|
| 403 |
+
"loss": 1.2762,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.09,
|
| 408 |
+
"eval_accuracy": 0.6061782240867615,
|
| 409 |
+
"eval_loss": 1.3001333475112915,
|
| 410 |
+
"eval_runtime": 1208.1233,
|
| 411 |
+
"eval_samples_per_second": 10.236,
|
| 412 |
+
"eval_steps_per_second": 5.118,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.13,
|
| 417 |
+
"learning_rate": 9.51112922951488e-05,
|
| 418 |
+
"loss": 1.3218,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.13,
|
| 423 |
+
"eval_accuracy": 0.592430830001831,
|
| 424 |
+
"eval_loss": 1.4698729515075684,
|
| 425 |
+
"eval_runtime": 1219.7567,
|
| 426 |
+
"eval_samples_per_second": 10.138,
|
| 427 |
+
"eval_steps_per_second": 5.069,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.17,
|
| 432 |
+
"learning_rate": 9.490786791683654e-05,
|
| 433 |
+
"loss": 1.3226,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.17,
|
| 438 |
+
"eval_accuracy": 0.5907326340675354,
|
| 439 |
+
"eval_loss": 1.396222710609436,
|
| 440 |
+
"eval_runtime": 1210.2892,
|
| 441 |
+
"eval_samples_per_second": 10.217,
|
| 442 |
+
"eval_steps_per_second": 5.109,
|
| 443 |
+
"step": 14500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.21,
|
| 447 |
+
"learning_rate": 9.470444353852426e-05,
|
| 448 |
+
"loss": 1.3306,
|
| 449 |
+
"step": 15000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.21,
|
| 453 |
+
"eval_accuracy": 0.6242924332618713,
|
| 454 |
+
"eval_loss": 1.3146133422851562,
|
| 455 |
+
"eval_runtime": 1215.13,
|
| 456 |
+
"eval_samples_per_second": 10.177,
|
| 457 |
+
"eval_steps_per_second": 5.088,
|
| 458 |
+
"step": 15000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.25,
|
| 462 |
+
"learning_rate": 9.45006114961272e-05,
|
| 463 |
+
"loss": 1.307,
|
| 464 |
+
"step": 15500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.25,
|
| 468 |
+
"eval_accuracy": 0.6107068061828613,
|
| 469 |
+
"eval_loss": 1.3835194110870361,
|
| 470 |
+
"eval_runtime": 1216.4262,
|
| 471 |
+
"eval_samples_per_second": 10.166,
|
| 472 |
+
"eval_steps_per_second": 5.083,
|
| 473 |
+
"step": 15500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.29,
|
| 477 |
+
"learning_rate": 9.429677945373014e-05,
|
| 478 |
+
"loss": 1.239,
|
| 479 |
+
"step": 16000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.29,
|
| 483 |
+
"eval_accuracy": 0.6234028935432434,
|
| 484 |
+
"eval_loss": 1.333041787147522,
|
| 485 |
+
"eval_runtime": 1214.4499,
|
| 486 |
+
"eval_samples_per_second": 10.182,
|
| 487 |
+
"eval_steps_per_second": 5.091,
|
| 488 |
+
"step": 16000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.33,
|
| 492 |
+
"learning_rate": 9.409294741133307e-05,
|
| 493 |
+
"loss": 1.2486,
|
| 494 |
+
"step": 16500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.33,
|
| 498 |
+
"eval_accuracy": 0.644670844078064,
|
| 499 |
+
"eval_loss": 1.187103033065796,
|
| 500 |
+
"eval_runtime": 1217.2373,
|
| 501 |
+
"eval_samples_per_second": 10.159,
|
| 502 |
+
"eval_steps_per_second": 5.08,
|
| 503 |
+
"step": 16500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 1.37,
|
| 507 |
+
"learning_rate": 9.3889115368936e-05,
|
| 508 |
+
"loss": 1.2305,
|
| 509 |
+
"step": 17000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.37,
|
| 513 |
+
"eval_accuracy": 0.6330260634422302,
|
| 514 |
+
"eval_loss": 1.280613660812378,
|
| 515 |
+
"eval_runtime": 1218.1968,
|
| 516 |
+
"eval_samples_per_second": 10.151,
|
| 517 |
+
"eval_steps_per_second": 5.076,
|
| 518 |
+
"step": 17000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.42,
|
| 522 |
+
"learning_rate": 9.368528332653894e-05,
|
| 523 |
+
"loss": 1.2427,
|
| 524 |
+
"step": 17500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.42,
|
| 528 |
+
"eval_accuracy": 0.6369885206222534,
|
| 529 |
+
"eval_loss": 1.3734965324401855,
|
| 530 |
+
"eval_runtime": 1216.3233,
|
| 531 |
+
"eval_samples_per_second": 10.167,
|
| 532 |
+
"eval_steps_per_second": 5.083,
|
| 533 |
+
"step": 17500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.46,
|
| 537 |
+
"learning_rate": 9.348145128414187e-05,
|
| 538 |
+
"loss": 1.232,
|
| 539 |
+
"step": 18000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.46,
|
| 543 |
+
"eval_accuracy": 0.6478247046470642,
|
| 544 |
+
"eval_loss": 1.244884729385376,
|
| 545 |
+
"eval_runtime": 1226.612,
|
| 546 |
+
"eval_samples_per_second": 10.081,
|
| 547 |
+
"eval_steps_per_second": 5.041,
|
| 548 |
+
"step": 18000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.5,
|
| 552 |
+
"learning_rate": 9.32776192417448e-05,
|
| 553 |
+
"loss": 1.2819,
|
| 554 |
+
"step": 18500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 1.5,
|
| 558 |
+
"eval_accuracy": 0.6360180974006653,
|
| 559 |
+
"eval_loss": 1.2376329898834229,
|
| 560 |
+
"eval_runtime": 1221.3054,
|
| 561 |
+
"eval_samples_per_second": 10.125,
|
| 562 |
+
"eval_steps_per_second": 5.063,
|
| 563 |
+
"step": 18500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.54,
|
| 567 |
+
"learning_rate": 9.307378719934773e-05,
|
| 568 |
+
"loss": 1.206,
|
| 569 |
+
"step": 19000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.54,
|
| 573 |
+
"eval_accuracy": 0.6438621878623962,
|
| 574 |
+
"eval_loss": 1.3146148920059204,
|
| 575 |
+
"eval_runtime": 1216.9126,
|
| 576 |
+
"eval_samples_per_second": 10.162,
|
| 577 |
+
"eval_steps_per_second": 5.081,
|
| 578 |
+
"step": 19000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.58,
|
| 582 |
+
"learning_rate": 9.286995515695067e-05,
|
| 583 |
+
"loss": 1.1787,
|
| 584 |
+
"step": 19500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.58,
|
| 588 |
+
"eval_accuracy": 0.6048035025596619,
|
| 589 |
+
"eval_loss": 1.5405412912368774,
|
| 590 |
+
"eval_runtime": 1222.2408,
|
| 591 |
+
"eval_samples_per_second": 10.117,
|
| 592 |
+
"eval_steps_per_second": 5.059,
|
| 593 |
+
"step": 19500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 1.62,
|
| 597 |
+
"learning_rate": 9.266612311455361e-05,
|
| 598 |
+
"loss": 1.2037,
|
| 599 |
+
"step": 20000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.62,
|
| 603 |
+
"eval_accuracy": 0.6377162933349609,
|
| 604 |
+
"eval_loss": 1.245172142982483,
|
| 605 |
+
"eval_runtime": 1222.2415,
|
| 606 |
+
"eval_samples_per_second": 10.117,
|
| 607 |
+
"eval_steps_per_second": 5.059,
|
| 608 |
+
"step": 20000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.66,
|
| 612 |
+
"learning_rate": 9.246269873624134e-05,
|
| 613 |
+
"loss": 1.2359,
|
| 614 |
+
"step": 20500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 1.66,
|
| 618 |
+
"eval_accuracy": 0.6452369689941406,
|
| 619 |
+
"eval_loss": 1.305955171585083,
|
| 620 |
+
"eval_runtime": 1226.2844,
|
| 621 |
+
"eval_samples_per_second": 10.084,
|
| 622 |
+
"eval_steps_per_second": 5.042,
|
| 623 |
+
"step": 20500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.7,
|
| 627 |
+
"learning_rate": 9.225886669384428e-05,
|
| 628 |
+
"loss": 1.258,
|
| 629 |
+
"step": 21000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.7,
|
| 633 |
+
"eval_accuracy": 0.6490376591682434,
|
| 634 |
+
"eval_loss": 1.2557748556137085,
|
| 635 |
+
"eval_runtime": 1225.4315,
|
| 636 |
+
"eval_samples_per_second": 10.091,
|
| 637 |
+
"eval_steps_per_second": 5.046,
|
| 638 |
+
"step": 21000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 1.74,
|
| 642 |
+
"learning_rate": 9.2055442315532e-05,
|
| 643 |
+
"loss": 1.124,
|
| 644 |
+
"step": 21500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.74,
|
| 648 |
+
"eval_accuracy": 0.6577712893486023,
|
| 649 |
+
"eval_loss": 1.201909065246582,
|
| 650 |
+
"eval_runtime": 1234.6035,
|
| 651 |
+
"eval_samples_per_second": 10.016,
|
| 652 |
+
"eval_steps_per_second": 5.008,
|
| 653 |
+
"step": 21500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.78,
|
| 657 |
+
"learning_rate": 9.185201793721973e-05,
|
| 658 |
+
"loss": 1.2008,
|
| 659 |
+
"step": 22000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 1.78,
|
| 663 |
+
"eval_accuracy": 0.6518680453300476,
|
| 664 |
+
"eval_loss": 1.2547849416732788,
|
| 665 |
+
"eval_runtime": 1221.6378,
|
| 666 |
+
"eval_samples_per_second": 10.122,
|
| 667 |
+
"eval_steps_per_second": 5.061,
|
| 668 |
+
"step": 22000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.82,
|
| 672 |
+
"learning_rate": 9.164818589482268e-05,
|
| 673 |
+
"loss": 1.231,
|
| 674 |
+
"step": 22500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.82,
|
| 678 |
+
"eval_accuracy": 0.6403040885925293,
|
| 679 |
+
"eval_loss": 1.2401455640792847,
|
| 680 |
+
"eval_runtime": 1227.1884,
|
| 681 |
+
"eval_samples_per_second": 10.077,
|
| 682 |
+
"eval_steps_per_second": 5.038,
|
| 683 |
+
"step": 22500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.86,
|
| 687 |
+
"learning_rate": 9.14443538524256e-05,
|
| 688 |
+
"loss": 1.176,
|
| 689 |
+
"step": 23000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.86,
|
| 693 |
+
"eval_accuracy": 0.630195677280426,
|
| 694 |
+
"eval_loss": 1.3183224201202393,
|
| 695 |
+
"eval_runtime": 1230.2141,
|
| 696 |
+
"eval_samples_per_second": 10.052,
|
| 697 |
+
"eval_steps_per_second": 5.026,
|
| 698 |
+
"step": 23000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.9,
|
| 702 |
+
"learning_rate": 9.124052181002855e-05,
|
| 703 |
+
"loss": 1.1756,
|
| 704 |
+
"step": 23500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.9,
|
| 708 |
+
"eval_accuracy": 0.6433770060539246,
|
| 709 |
+
"eval_loss": 1.3021215200424194,
|
| 710 |
+
"eval_runtime": 1225.278,
|
| 711 |
+
"eval_samples_per_second": 10.092,
|
| 712 |
+
"eval_steps_per_second": 5.046,
|
| 713 |
+
"step": 23500
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.94,
|
| 717 |
+
"learning_rate": 9.103709743171626e-05,
|
| 718 |
+
"loss": 1.2213,
|
| 719 |
+
"step": 24000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.94,
|
| 723 |
+
"eval_accuracy": 0.6450752019882202,
|
| 724 |
+
"eval_loss": 1.2812703847885132,
|
| 725 |
+
"eval_runtime": 1226.6055,
|
| 726 |
+
"eval_samples_per_second": 10.081,
|
| 727 |
+
"eval_steps_per_second": 5.041,
|
| 728 |
+
"step": 24000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.98,
|
| 732 |
+
"learning_rate": 9.08332653893192e-05,
|
| 733 |
+
"loss": 1.1343,
|
| 734 |
+
"step": 24500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.98,
|
| 738 |
+
"eval_accuracy": 0.6614103317260742,
|
| 739 |
+
"eval_loss": 1.2407350540161133,
|
| 740 |
+
"eval_runtime": 1227.7031,
|
| 741 |
+
"eval_samples_per_second": 10.072,
|
| 742 |
+
"eval_steps_per_second": 5.036,
|
| 743 |
+
"step": 24500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 2.02,
|
| 747 |
+
"learning_rate": 9.062943334692214e-05,
|
| 748 |
+
"loss": 1.1498,
|
| 749 |
+
"step": 25000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.02,
|
| 753 |
+
"eval_accuracy": 0.6570435166358948,
|
| 754 |
+
"eval_loss": 1.1911596059799194,
|
| 755 |
+
"eval_runtime": 1201.1782,
|
| 756 |
+
"eval_samples_per_second": 10.295,
|
| 757 |
+
"eval_steps_per_second": 5.147,
|
| 758 |
+
"step": 25000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.06,
|
| 762 |
+
"learning_rate": 9.042560130452508e-05,
|
| 763 |
+
"loss": 1.1015,
|
| 764 |
+
"step": 25500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.06,
|
| 768 |
+
"eval_accuracy": 0.6508976221084595,
|
| 769 |
+
"eval_loss": 1.360363245010376,
|
| 770 |
+
"eval_runtime": 1193.6402,
|
| 771 |
+
"eval_samples_per_second": 10.36,
|
| 772 |
+
"eval_steps_per_second": 5.18,
|
| 773 |
+
"step": 25500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.1,
|
| 777 |
+
"learning_rate": 9.022176926212801e-05,
|
| 778 |
+
"loss": 1.0794,
|
| 779 |
+
"step": 26000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.1,
|
| 783 |
+
"eval_accuracy": 0.6508976221084595,
|
| 784 |
+
"eval_loss": 1.3214635848999023,
|
| 785 |
+
"eval_runtime": 1206.1082,
|
| 786 |
+
"eval_samples_per_second": 10.253,
|
| 787 |
+
"eval_steps_per_second": 5.126,
|
| 788 |
+
"step": 26000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.14,
|
| 792 |
+
"learning_rate": 9.001793721973094e-05,
|
| 793 |
+
"loss": 1.0692,
|
| 794 |
+
"step": 26500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 2.14,
|
| 798 |
+
"eval_accuracy": 0.662542462348938,
|
| 799 |
+
"eval_loss": 1.2083770036697388,
|
| 800 |
+
"eval_runtime": 1204.5386,
|
| 801 |
+
"eval_samples_per_second": 10.266,
|
| 802 |
+
"eval_steps_per_second": 5.133,
|
| 803 |
+
"step": 26500
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.18,
|
| 807 |
+
"learning_rate": 8.981492050550348e-05,
|
| 808 |
+
"loss": 1.1552,
|
| 809 |
+
"step": 27000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.18,
|
| 813 |
+
"eval_accuracy": 0.6536471247673035,
|
| 814 |
+
"eval_loss": 1.2683591842651367,
|
| 815 |
+
"eval_runtime": 1207.0957,
|
| 816 |
+
"eval_samples_per_second": 10.244,
|
| 817 |
+
"eval_steps_per_second": 5.122,
|
| 818 |
+
"step": 27000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.22,
|
| 822 |
+
"learning_rate": 8.961108846310641e-05,
|
| 823 |
+
"loss": 1.0985,
|
| 824 |
+
"step": 27500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.22,
|
| 828 |
+
"eval_accuracy": 0.6645641326904297,
|
| 829 |
+
"eval_loss": 1.1616308689117432,
|
| 830 |
+
"eval_runtime": 1202.0246,
|
| 831 |
+
"eval_samples_per_second": 10.288,
|
| 832 |
+
"eval_steps_per_second": 5.144,
|
| 833 |
+
"step": 27500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 2.26,
|
| 837 |
+
"learning_rate": 8.940725642070934e-05,
|
| 838 |
+
"loss": 1.1239,
|
| 839 |
+
"step": 28000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.26,
|
| 843 |
+
"eval_accuracy": 0.669173538684845,
|
| 844 |
+
"eval_loss": 1.1362210512161255,
|
| 845 |
+
"eval_runtime": 1205.0904,
|
| 846 |
+
"eval_samples_per_second": 10.261,
|
| 847 |
+
"eval_steps_per_second": 5.131,
|
| 848 |
+
"step": 28000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.3,
|
| 852 |
+
"learning_rate": 8.920342437831227e-05,
|
| 853 |
+
"loss": 1.0899,
|
| 854 |
+
"step": 28500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.3,
|
| 858 |
+
"eval_accuracy": 0.6521106362342834,
|
| 859 |
+
"eval_loss": 1.3762954473495483,
|
| 860 |
+
"eval_runtime": 1210.3119,
|
| 861 |
+
"eval_samples_per_second": 10.217,
|
| 862 |
+
"eval_steps_per_second": 5.109,
|
| 863 |
+
"step": 28500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.35,
|
| 867 |
+
"learning_rate": 8.89995923359152e-05,
|
| 868 |
+
"loss": 1.0953,
|
| 869 |
+
"step": 29000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.35,
|
| 873 |
+
"eval_accuracy": 0.666747510433197,
|
| 874 |
+
"eval_loss": 1.1568596363067627,
|
| 875 |
+
"eval_runtime": 1213.4003,
|
| 876 |
+
"eval_samples_per_second": 10.191,
|
| 877 |
+
"eval_steps_per_second": 5.096,
|
| 878 |
+
"step": 29000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.39,
|
| 882 |
+
"learning_rate": 8.879576029351814e-05,
|
| 883 |
+
"loss": 1.1148,
|
| 884 |
+
"step": 29500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.39,
|
| 888 |
+
"eval_accuracy": 0.6541323065757751,
|
| 889 |
+
"eval_loss": 1.2405096292495728,
|
| 890 |
+
"eval_runtime": 1215.762,
|
| 891 |
+
"eval_samples_per_second": 10.171,
|
| 892 |
+
"eval_steps_per_second": 5.086,
|
| 893 |
+
"step": 29500
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.43,
|
| 897 |
+
"learning_rate": 8.859192825112108e-05,
|
| 898 |
+
"loss": 1.0715,
|
| 899 |
+
"step": 30000
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.43,
|
| 903 |
+
"eval_accuracy": 0.6622189879417419,
|
| 904 |
+
"eval_loss": 1.2203196287155151,
|
| 905 |
+
"eval_runtime": 1237.659,
|
| 906 |
+
"eval_samples_per_second": 9.991,
|
| 907 |
+
"eval_steps_per_second": 4.996,
|
| 908 |
+
"step": 30000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.47,
|
| 912 |
+
"learning_rate": 8.838809620872402e-05,
|
| 913 |
+
"loss": 1.1298,
|
| 914 |
+
"step": 30500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.47,
|
| 918 |
+
"eval_accuracy": 0.667960524559021,
|
| 919 |
+
"eval_loss": 1.258815884590149,
|
| 920 |
+
"eval_runtime": 1214.4985,
|
| 921 |
+
"eval_samples_per_second": 10.182,
|
| 922 |
+
"eval_steps_per_second": 5.091,
|
| 923 |
+
"step": 30500
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.51,
|
| 927 |
+
"learning_rate": 8.818426416632695e-05,
|
| 928 |
+
"loss": 1.1078,
|
| 929 |
+
"step": 31000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 2.51,
|
| 933 |
+
"eval_accuracy": 0.6598738431930542,
|
| 934 |
+
"eval_loss": 1.1994160413742065,
|
| 935 |
+
"eval_runtime": 1223.4805,
|
| 936 |
+
"eval_samples_per_second": 10.107,
|
| 937 |
+
"eval_steps_per_second": 5.054,
|
| 938 |
+
"step": 31000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 2.55,
|
| 942 |
+
"learning_rate": 8.79804321239299e-05,
|
| 943 |
+
"loss": 1.1229,
|
| 944 |
+
"step": 31500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 2.55,
|
| 948 |
+
"eval_accuracy": 0.6650493144989014,
|
| 949 |
+
"eval_loss": 1.1719839572906494,
|
| 950 |
+
"eval_runtime": 1217.7937,
|
| 951 |
+
"eval_samples_per_second": 10.154,
|
| 952 |
+
"eval_steps_per_second": 5.077,
|
| 953 |
+
"step": 31500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 2.59,
|
| 957 |
+
"learning_rate": 8.777660008153283e-05,
|
| 958 |
+
"loss": 1.1176,
|
| 959 |
+
"step": 32000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.59,
|
| 963 |
+
"eval_accuracy": 0.6727316975593567,
|
| 964 |
+
"eval_loss": 1.1364519596099854,
|
| 965 |
+
"eval_runtime": 1228.5963,
|
| 966 |
+
"eval_samples_per_second": 10.065,
|
| 967 |
+
"eval_steps_per_second": 5.033,
|
| 968 |
+
"step": 32000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.63,
|
| 972 |
+
"learning_rate": 8.757276803913576e-05,
|
| 973 |
+
"loss": 1.0393,
|
| 974 |
+
"step": 32500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 2.63,
|
| 978 |
+
"eval_accuracy": 0.6387676000595093,
|
| 979 |
+
"eval_loss": 1.356237769126892,
|
| 980 |
+
"eval_runtime": 1220.1605,
|
| 981 |
+
"eval_samples_per_second": 10.135,
|
| 982 |
+
"eval_steps_per_second": 5.067,
|
| 983 |
+
"step": 32500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.67,
|
| 987 |
+
"learning_rate": 8.736893599673869e-05,
|
| 988 |
+
"loss": 1.0686,
|
| 989 |
+
"step": 33000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.67,
|
| 993 |
+
"eval_accuracy": 0.6770985126495361,
|
| 994 |
+
"eval_loss": 1.2882039546966553,
|
| 995 |
+
"eval_runtime": 1227.7272,
|
| 996 |
+
"eval_samples_per_second": 10.072,
|
| 997 |
+
"eval_steps_per_second": 5.036,
|
| 998 |
+
"step": 33000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.71,
|
| 1002 |
+
"learning_rate": 8.716591928251121e-05,
|
| 1003 |
+
"loss": 1.0984,
|
| 1004 |
+
"step": 33500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 2.71,
|
| 1008 |
+
"eval_accuracy": 0.6504932641983032,
|
| 1009 |
+
"eval_loss": 1.2953248023986816,
|
| 1010 |
+
"eval_runtime": 1228.2391,
|
| 1011 |
+
"eval_samples_per_second": 10.068,
|
| 1012 |
+
"eval_steps_per_second": 5.034,
|
| 1013 |
+
"step": 33500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.75,
|
| 1017 |
+
"learning_rate": 8.696208724011415e-05,
|
| 1018 |
+
"loss": 1.0653,
|
| 1019 |
+
"step": 34000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.75,
|
| 1023 |
+
"eval_accuracy": 0.6635128855705261,
|
| 1024 |
+
"eval_loss": 1.2924034595489502,
|
| 1025 |
+
"eval_runtime": 1226.8106,
|
| 1026 |
+
"eval_samples_per_second": 10.08,
|
| 1027 |
+
"eval_steps_per_second": 5.04,
|
| 1028 |
+
"step": 34000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.79,
|
| 1032 |
+
"learning_rate": 8.675825519771708e-05,
|
| 1033 |
+
"loss": 1.0878,
|
| 1034 |
+
"step": 34500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 2.79,
|
| 1038 |
+
"eval_accuracy": 0.6699013710021973,
|
| 1039 |
+
"eval_loss": 1.2249584197998047,
|
| 1040 |
+
"eval_runtime": 1223.8019,
|
| 1041 |
+
"eval_samples_per_second": 10.105,
|
| 1042 |
+
"eval_steps_per_second": 5.052,
|
| 1043 |
+
"step": 34500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.83,
|
| 1047 |
+
"learning_rate": 8.655442315532001e-05,
|
| 1048 |
+
"loss": 1.0446,
|
| 1049 |
+
"step": 35000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.83,
|
| 1053 |
+
"eval_accuracy": 0.6664240956306458,
|
| 1054 |
+
"eval_loss": 1.3492021560668945,
|
| 1055 |
+
"eval_runtime": 1227.92,
|
| 1056 |
+
"eval_samples_per_second": 10.071,
|
| 1057 |
+
"eval_steps_per_second": 5.035,
|
| 1058 |
+
"step": 35000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.87,
|
| 1062 |
+
"learning_rate": 8.635059111292296e-05,
|
| 1063 |
+
"loss": 1.0754,
|
| 1064 |
+
"step": 35500
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 2.87,
|
| 1068 |
+
"eval_accuracy": 0.6584182381629944,
|
| 1069 |
+
"eval_loss": 1.275933861732483,
|
| 1070 |
+
"eval_runtime": 1225.9895,
|
| 1071 |
+
"eval_samples_per_second": 10.087,
|
| 1072 |
+
"eval_steps_per_second": 5.043,
|
| 1073 |
+
"step": 35500
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.91,
|
| 1077 |
+
"learning_rate": 8.614675907052589e-05,
|
| 1078 |
+
"loss": 1.0441,
|
| 1079 |
+
"step": 36000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.91,
|
| 1083 |
+
"eval_accuracy": 0.6821931004524231,
|
| 1084 |
+
"eval_loss": 1.1557066440582275,
|
| 1085 |
+
"eval_runtime": 1227.8238,
|
| 1086 |
+
"eval_samples_per_second": 10.071,
|
| 1087 |
+
"eval_steps_per_second": 5.036,
|
| 1088 |
+
"step": 36000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.95,
|
| 1092 |
+
"learning_rate": 8.594374235629842e-05,
|
| 1093 |
+
"loss": 1.2804,
|
| 1094 |
+
"step": 36500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.95,
|
| 1098 |
+
"eval_accuracy": 0.6013262271881104,
|
| 1099 |
+
"eval_loss": 1.4437016248703003,
|
| 1100 |
+
"eval_runtime": 1228.8358,
|
| 1101 |
+
"eval_samples_per_second": 10.063,
|
| 1102 |
+
"eval_steps_per_second": 5.032,
|
| 1103 |
+
"step": 36500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.99,
|
| 1107 |
+
"learning_rate": 8.573991031390135e-05,
|
| 1108 |
+
"loss": 1.0833,
|
| 1109 |
+
"step": 37000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.99,
|
| 1113 |
+
"eval_accuracy": 0.6906032562255859,
|
| 1114 |
+
"eval_loss": 1.1393189430236816,
|
| 1115 |
+
"eval_runtime": 1227.4427,
|
| 1116 |
+
"eval_samples_per_second": 10.075,
|
| 1117 |
+
"eval_steps_per_second": 5.037,
|
| 1118 |
+
"step": 37000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 3.03,
|
| 1122 |
+
"learning_rate": 8.553607827150428e-05,
|
| 1123 |
+
"loss": 0.9851,
|
| 1124 |
+
"step": 37500
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 3.03,
|
| 1128 |
+
"eval_accuracy": 0.6836487054824829,
|
| 1129 |
+
"eval_loss": 1.1196837425231934,
|
| 1130 |
+
"eval_runtime": 1223.1266,
|
| 1131 |
+
"eval_samples_per_second": 10.11,
|
| 1132 |
+
"eval_steps_per_second": 5.055,
|
| 1133 |
+
"step": 37500
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 3.07,
|
| 1137 |
+
"learning_rate": 8.533224622910721e-05,
|
| 1138 |
+
"loss": 0.95,
|
| 1139 |
+
"step": 38000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 3.07,
|
| 1143 |
+
"eval_accuracy": 0.6299530863761902,
|
| 1144 |
+
"eval_loss": 1.4798767566680908,
|
| 1145 |
+
"eval_runtime": 1224.7327,
|
| 1146 |
+
"eval_samples_per_second": 10.097,
|
| 1147 |
+
"eval_steps_per_second": 5.048,
|
| 1148 |
+
"step": 38000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 3.11,
|
| 1152 |
+
"learning_rate": 8.512841418671016e-05,
|
| 1153 |
+
"loss": 1.0258,
|
| 1154 |
+
"step": 38500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 3.11,
|
| 1158 |
+
"eval_accuracy": 0.6828400492668152,
|
| 1159 |
+
"eval_loss": 1.1659743785858154,
|
| 1160 |
+
"eval_runtime": 1235.0907,
|
| 1161 |
+
"eval_samples_per_second": 10.012,
|
| 1162 |
+
"eval_steps_per_second": 5.006,
|
| 1163 |
+
"step": 38500
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 3.15,
|
| 1167 |
+
"learning_rate": 8.492458214431309e-05,
|
| 1168 |
+
"loss": 1.0604,
|
| 1169 |
+
"step": 39000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 3.15,
|
| 1173 |
+
"eval_accuracy": 0.6709526181221008,
|
| 1174 |
+
"eval_loss": 1.2192966938018799,
|
| 1175 |
+
"eval_runtime": 1228.4696,
|
| 1176 |
+
"eval_samples_per_second": 10.066,
|
| 1177 |
+
"eval_steps_per_second": 5.033,
|
| 1178 |
+
"step": 39000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.19,
|
| 1182 |
+
"learning_rate": 8.472075010191602e-05,
|
| 1183 |
+
"loss": 1.0496,
|
| 1184 |
+
"step": 39500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 3.19,
|
| 1188 |
+
"eval_accuracy": 0.6782306432723999,
|
| 1189 |
+
"eval_loss": 1.2562731504440308,
|
| 1190 |
+
"eval_runtime": 1229.0843,
|
| 1191 |
+
"eval_samples_per_second": 10.061,
|
| 1192 |
+
"eval_steps_per_second": 5.031,
|
| 1193 |
+
"step": 39500
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 3.23,
|
| 1197 |
+
"learning_rate": 8.451732572360376e-05,
|
| 1198 |
+
"loss": 2.5985,
|
| 1199 |
+
"step": 40000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.23,
|
| 1203 |
+
"eval_accuracy": 0.1606016457080841,
|
| 1204 |
+
"eval_loss": 2.8276100158691406,
|
| 1205 |
+
"eval_runtime": 1227.4333,
|
| 1206 |
+
"eval_samples_per_second": 10.075,
|
| 1207 |
+
"eval_steps_per_second": 5.037,
|
| 1208 |
+
"step": 40000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 3.28,
|
| 1212 |
+
"learning_rate": 8.431430900937628e-05,
|
| 1213 |
+
"loss": 2.5628,
|
| 1214 |
+
"step": 40500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.28,
|
| 1218 |
+
"eval_accuracy": 0.6409509778022766,
|
| 1219 |
+
"eval_loss": 1.2782047986984253,
|
| 1220 |
+
"eval_runtime": 1226.8685,
|
| 1221 |
+
"eval_samples_per_second": 10.079,
|
| 1222 |
+
"eval_steps_per_second": 5.04,
|
| 1223 |
+
"step": 40500
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 3.32,
|
| 1227 |
+
"learning_rate": 8.411047696697921e-05,
|
| 1228 |
+
"loss": 1.0363,
|
| 1229 |
+
"step": 41000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 3.32,
|
| 1233 |
+
"eval_accuracy": 0.6780688762664795,
|
| 1234 |
+
"eval_loss": 1.2363730669021606,
|
| 1235 |
+
"eval_runtime": 1229.116,
|
| 1236 |
+
"eval_samples_per_second": 10.061,
|
| 1237 |
+
"eval_steps_per_second": 5.03,
|
| 1238 |
+
"step": 41000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 3.36,
|
| 1242 |
+
"learning_rate": 8.390664492458214e-05,
|
| 1243 |
+
"loss": 1.0159,
|
| 1244 |
+
"step": 41500
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 3.36,
|
| 1248 |
+
"eval_accuracy": 0.6617338061332703,
|
| 1249 |
+
"eval_loss": 1.323940396308899,
|
| 1250 |
+
"eval_runtime": 1225.656,
|
| 1251 |
+
"eval_samples_per_second": 10.089,
|
| 1252 |
+
"eval_steps_per_second": 5.045,
|
| 1253 |
+
"step": 41500
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 3.4,
|
| 1257 |
+
"learning_rate": 8.370281288218509e-05,
|
| 1258 |
+
"loss": 1.0396,
|
| 1259 |
+
"step": 42000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 3.4,
|
| 1263 |
+
"eval_accuracy": 0.6728934049606323,
|
| 1264 |
+
"eval_loss": 1.187890887260437,
|
| 1265 |
+
"eval_runtime": 1229.8463,
|
| 1266 |
+
"eval_samples_per_second": 10.055,
|
| 1267 |
+
"eval_steps_per_second": 5.027,
|
| 1268 |
+
"step": 42000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 3.44,
|
| 1272 |
+
"learning_rate": 8.349898083978802e-05,
|
| 1273 |
+
"loss": 1.0262,
|
| 1274 |
+
"step": 42500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 3.44,
|
| 1278 |
+
"eval_accuracy": 0.6631894111633301,
|
| 1279 |
+
"eval_loss": 1.18537437915802,
|
| 1280 |
+
"eval_runtime": 1238.5652,
|
| 1281 |
+
"eval_samples_per_second": 9.984,
|
| 1282 |
+
"eval_steps_per_second": 4.992,
|
| 1283 |
+
"step": 42500
|
| 1284 |
+
}
|
| 1285 |
+
],
|
| 1286 |
+
"max_steps": 247300,
|
| 1287 |
+
"num_train_epochs": 20,
|
| 1288 |
+
"total_flos": 2.8496051091712414e+19,
|
| 1289 |
+
"trial_name": null,
|
| 1290 |
+
"trial_params": null
|
| 1291 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fcddc3abd62ffa4c4f7e0ca36ff6f4c75c873f2ecad8ec229fdda0ca3fbd3cbd
|
| 3 |
+
size 3512
|