Anders L|hr commited on
Commit ·
d4b56e8
1
Parent(s): a196bef
Best 5 class model
Browse files- HuggingfaceBest5ClassModel +1 -0
- checkpoint-5432/HuggingfaceBest5ClassModel +1 -0
- checkpoint-5432/config.json +124 -0
- checkpoint-5432/model.safetensors +3 -0
- checkpoint-5432/optimizer.pt +3 -0
- checkpoint-5432/preprocessor_config.json +9 -0
- checkpoint-5432/rng_state.pth +3 -0
- checkpoint-5432/scheduler.pt +3 -0
- checkpoint-5432/trainer_state.json +2793 -0
- checkpoint-5432/training_args.bin +3 -0
- config.json +124 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2793 -0
- training_args.bin +3 -0
HuggingfaceBest5ClassModel
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
Subproject commit a196bef4dd25c395f33565fcc21a540eae728a16
|
checkpoint-5432/HuggingfaceBest5ClassModel
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
Subproject commit a196bef4dd25c395f33565fcc21a540eae728a16
|
checkpoint-5432/config.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test",
|
| 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 |
+
"Wav2Vec2ForSequenceClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 128,
|
| 15 |
+
"codevector_dim": 128,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
256,
|
| 20 |
+
256,
|
| 21 |
+
256,
|
| 22 |
+
256,
|
| 23 |
+
256,
|
| 24 |
+
256,
|
| 25 |
+
256
|
| 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": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "audio-classification",
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_dropout_prob": 0.0,
|
| 60 |
+
"hidden_size": 384,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "Helicopter",
|
| 63 |
+
"1": "Jet",
|
| 64 |
+
"2": "Racecar",
|
| 65 |
+
"3": "Rail",
|
| 66 |
+
"4": "Truck"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 1536,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"Helicopter": "0",
|
| 72 |
+
"Jet": "1",
|
| 73 |
+
"Racecar": "2",
|
| 74 |
+
"Rail": "3",
|
| 75 |
+
"Truck": "4"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.65,
|
| 85 |
+
"model_type": "wav2vec2",
|
| 86 |
+
"num_adapter_layers": 3,
|
| 87 |
+
"num_attention_heads": 6,
|
| 88 |
+
"num_codevector_groups": 2,
|
| 89 |
+
"num_codevectors_per_group": 320,
|
| 90 |
+
"num_conv_pos_embedding_groups": 16,
|
| 91 |
+
"num_conv_pos_embeddings": 128,
|
| 92 |
+
"num_feat_extract_layers": 7,
|
| 93 |
+
"num_hidden_layers": 6,
|
| 94 |
+
"num_negatives": 100,
|
| 95 |
+
"output_hidden_size": 384,
|
| 96 |
+
"pad_token_id": 0,
|
| 97 |
+
"proj_codevector_dim": 128,
|
| 98 |
+
"tdnn_dilation": [
|
| 99 |
+
1,
|
| 100 |
+
2,
|
| 101 |
+
3,
|
| 102 |
+
1,
|
| 103 |
+
1
|
| 104 |
+
],
|
| 105 |
+
"tdnn_dim": [
|
| 106 |
+
512,
|
| 107 |
+
512,
|
| 108 |
+
512,
|
| 109 |
+
512,
|
| 110 |
+
1500
|
| 111 |
+
],
|
| 112 |
+
"tdnn_kernel": [
|
| 113 |
+
5,
|
| 114 |
+
3,
|
| 115 |
+
3,
|
| 116 |
+
1,
|
| 117 |
+
1
|
| 118 |
+
],
|
| 119 |
+
"torch_dtype": "float32",
|
| 120 |
+
"transformers_version": "4.39.0.dev0",
|
| 121 |
+
"use_weighted_layer_sum": false,
|
| 122 |
+
"vocab_size": 32,
|
| 123 |
+
"xvector_output_dim": 512
|
| 124 |
+
}
|
checkpoint-5432/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12b72afca351d5838b740f0ec6003c2e1e2a8c0f5156e6629ad3e2ef735bb540
|
| 3 |
+
size 52151348
|
checkpoint-5432/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd83ae850a6f38e955a9a0ecf3728e21f005733ae6ae7f948544118441a4714b
|
| 3 |
+
size 95909946
|
checkpoint-5432/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": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-5432/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0524d4c5bb50cb3a888e246202453f6e7f310c8e7d978c2791811f64716d2d2c
|
| 3 |
+
size 14244
|
checkpoint-5432/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66b1ff4c92709ee514d60150ad1c67a13e001dac71642548c74f443c1156d358
|
| 3 |
+
size 1064
|
checkpoint-5432/trainer_state.json
ADDED
|
@@ -0,0 +1,2793 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7937062937062938,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-5Class-train-test-finetune/checkpoint-4122",
|
| 4 |
+
"epoch": 224.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5432,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.99,
|
| 13 |
+
"eval_accuracy": 0.34265734265734266,
|
| 14 |
+
"eval_loss": 1.5984586477279663,
|
| 15 |
+
"eval_runtime": 5.3437,
|
| 16 |
+
"eval_samples_per_second": 53.521,
|
| 17 |
+
"eval_steps_per_second": 3.368,
|
| 18 |
+
"step": 24
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.98,
|
| 22 |
+
"eval_accuracy": 0.33916083916083917,
|
| 23 |
+
"eval_loss": 1.5969289541244507,
|
| 24 |
+
"eval_runtime": 3.8653,
|
| 25 |
+
"eval_samples_per_second": 73.992,
|
| 26 |
+
"eval_steps_per_second": 4.657,
|
| 27 |
+
"step": 48
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.06,
|
| 31 |
+
"grad_norm": 1.0544973611831665,
|
| 32 |
+
"learning_rate": 2.4999999999999998e-06,
|
| 33 |
+
"loss": 1.5969,
|
| 34 |
+
"step": 50
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.97,
|
| 38 |
+
"eval_accuracy": 0.32867132867132864,
|
| 39 |
+
"eval_loss": 1.5943816900253296,
|
| 40 |
+
"eval_runtime": 6.1748,
|
| 41 |
+
"eval_samples_per_second": 46.317,
|
| 42 |
+
"eval_steps_per_second": 2.915,
|
| 43 |
+
"step": 72
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.3146853146853147,
|
| 48 |
+
"eval_loss": 1.5906767845153809,
|
| 49 |
+
"eval_runtime": 5.1678,
|
| 50 |
+
"eval_samples_per_second": 55.343,
|
| 51 |
+
"eval_steps_per_second": 3.483,
|
| 52 |
+
"step": 97
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.12,
|
| 56 |
+
"grad_norm": 0.8443157076835632,
|
| 57 |
+
"learning_rate": 4.9999999999999996e-06,
|
| 58 |
+
"loss": 1.5896,
|
| 59 |
+
"step": 100
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.99,
|
| 63 |
+
"eval_accuracy": 0.2972027972027972,
|
| 64 |
+
"eval_loss": 1.5860023498535156,
|
| 65 |
+
"eval_runtime": 4.9416,
|
| 66 |
+
"eval_samples_per_second": 57.876,
|
| 67 |
+
"eval_steps_per_second": 3.643,
|
| 68 |
+
"step": 121
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.98,
|
| 72 |
+
"eval_accuracy": 0.2692307692307692,
|
| 73 |
+
"eval_loss": 1.5806005001068115,
|
| 74 |
+
"eval_runtime": 4.1837,
|
| 75 |
+
"eval_samples_per_second": 68.36,
|
| 76 |
+
"eval_steps_per_second": 4.302,
|
| 77 |
+
"step": 145
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.19,
|
| 81 |
+
"grad_norm": 1.0938074588775635,
|
| 82 |
+
"learning_rate": 7.5e-06,
|
| 83 |
+
"loss": 1.5743,
|
| 84 |
+
"step": 150
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.97,
|
| 88 |
+
"eval_accuracy": 0.25874125874125875,
|
| 89 |
+
"eval_loss": 1.5742768049240112,
|
| 90 |
+
"eval_runtime": 7.1914,
|
| 91 |
+
"eval_samples_per_second": 39.77,
|
| 92 |
+
"eval_steps_per_second": 2.503,
|
| 93 |
+
"step": 169
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.0,
|
| 97 |
+
"eval_accuracy": 0.23426573426573427,
|
| 98 |
+
"eval_loss": 1.5664165019989014,
|
| 99 |
+
"eval_runtime": 5.6489,
|
| 100 |
+
"eval_samples_per_second": 50.629,
|
| 101 |
+
"eval_steps_per_second": 3.186,
|
| 102 |
+
"step": 194
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.25,
|
| 106 |
+
"grad_norm": 0.9692079424858093,
|
| 107 |
+
"learning_rate": 9.999999999999999e-06,
|
| 108 |
+
"loss": 1.5508,
|
| 109 |
+
"step": 200
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 8.99,
|
| 113 |
+
"eval_accuracy": 0.22727272727272727,
|
| 114 |
+
"eval_loss": 1.557572841644287,
|
| 115 |
+
"eval_runtime": 5.5182,
|
| 116 |
+
"eval_samples_per_second": 51.828,
|
| 117 |
+
"eval_steps_per_second": 3.262,
|
| 118 |
+
"step": 218
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 9.98,
|
| 122 |
+
"eval_accuracy": 0.22727272727272727,
|
| 123 |
+
"eval_loss": 1.5482373237609863,
|
| 124 |
+
"eval_runtime": 5.3205,
|
| 125 |
+
"eval_samples_per_second": 53.754,
|
| 126 |
+
"eval_steps_per_second": 3.383,
|
| 127 |
+
"step": 242
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 10.31,
|
| 131 |
+
"grad_norm": 1.02046799659729,
|
| 132 |
+
"learning_rate": 1.25e-05,
|
| 133 |
+
"loss": 1.5157,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.97,
|
| 138 |
+
"eval_accuracy": 0.22727272727272727,
|
| 139 |
+
"eval_loss": 1.539355993270874,
|
| 140 |
+
"eval_runtime": 6.3116,
|
| 141 |
+
"eval_samples_per_second": 45.313,
|
| 142 |
+
"eval_steps_per_second": 2.852,
|
| 143 |
+
"step": 266
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 12.0,
|
| 147 |
+
"eval_accuracy": 0.22727272727272727,
|
| 148 |
+
"eval_loss": 1.5350520610809326,
|
| 149 |
+
"eval_runtime": 4.3422,
|
| 150 |
+
"eval_samples_per_second": 65.865,
|
| 151 |
+
"eval_steps_per_second": 4.145,
|
| 152 |
+
"step": 291
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 12.37,
|
| 156 |
+
"grad_norm": 1.6058833599090576,
|
| 157 |
+
"learning_rate": 1.5e-05,
|
| 158 |
+
"loss": 1.4534,
|
| 159 |
+
"step": 300
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 12.99,
|
| 163 |
+
"eval_accuracy": 0.22727272727272727,
|
| 164 |
+
"eval_loss": 1.5525730848312378,
|
| 165 |
+
"eval_runtime": 5.245,
|
| 166 |
+
"eval_samples_per_second": 54.528,
|
| 167 |
+
"eval_steps_per_second": 3.432,
|
| 168 |
+
"step": 315
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 13.98,
|
| 172 |
+
"eval_accuracy": 0.22727272727272727,
|
| 173 |
+
"eval_loss": 1.599926471710205,
|
| 174 |
+
"eval_runtime": 6.0088,
|
| 175 |
+
"eval_samples_per_second": 47.597,
|
| 176 |
+
"eval_steps_per_second": 2.996,
|
| 177 |
+
"step": 339
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 14.43,
|
| 181 |
+
"grad_norm": 0.8243080377578735,
|
| 182 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 183 |
+
"loss": 1.3638,
|
| 184 |
+
"step": 350
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 14.97,
|
| 188 |
+
"eval_accuracy": 0.22727272727272727,
|
| 189 |
+
"eval_loss": 1.5896875858306885,
|
| 190 |
+
"eval_runtime": 4.8752,
|
| 191 |
+
"eval_samples_per_second": 58.664,
|
| 192 |
+
"eval_steps_per_second": 3.692,
|
| 193 |
+
"step": 363
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 16.0,
|
| 197 |
+
"eval_accuracy": 0.26573426573426573,
|
| 198 |
+
"eval_loss": 1.560091495513916,
|
| 199 |
+
"eval_runtime": 5.5082,
|
| 200 |
+
"eval_samples_per_second": 51.922,
|
| 201 |
+
"eval_steps_per_second": 3.268,
|
| 202 |
+
"step": 388
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 16.49,
|
| 206 |
+
"grad_norm": 0.7977257370948792,
|
| 207 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 208 |
+
"loss": 1.2951,
|
| 209 |
+
"step": 400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 16.99,
|
| 213 |
+
"eval_accuracy": 0.2937062937062937,
|
| 214 |
+
"eval_loss": 1.5349317789077759,
|
| 215 |
+
"eval_runtime": 4.7526,
|
| 216 |
+
"eval_samples_per_second": 60.178,
|
| 217 |
+
"eval_steps_per_second": 3.787,
|
| 218 |
+
"step": 412
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 17.98,
|
| 222 |
+
"eval_accuracy": 0.34265734265734266,
|
| 223 |
+
"eval_loss": 1.5053907632827759,
|
| 224 |
+
"eval_runtime": 4.8638,
|
| 225 |
+
"eval_samples_per_second": 58.801,
|
| 226 |
+
"eval_steps_per_second": 3.701,
|
| 227 |
+
"step": 436
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 18.56,
|
| 231 |
+
"grad_norm": 0.7064552903175354,
|
| 232 |
+
"learning_rate": 2.25e-05,
|
| 233 |
+
"loss": 1.2369,
|
| 234 |
+
"step": 450
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 18.97,
|
| 238 |
+
"eval_accuracy": 0.3741258741258741,
|
| 239 |
+
"eval_loss": 1.4689087867736816,
|
| 240 |
+
"eval_runtime": 4.3712,
|
| 241 |
+
"eval_samples_per_second": 65.428,
|
| 242 |
+
"eval_steps_per_second": 4.118,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 20.0,
|
| 247 |
+
"eval_accuracy": 0.4370629370629371,
|
| 248 |
+
"eval_loss": 1.404613971710205,
|
| 249 |
+
"eval_runtime": 4.7203,
|
| 250 |
+
"eval_samples_per_second": 60.59,
|
| 251 |
+
"eval_steps_per_second": 3.813,
|
| 252 |
+
"step": 485
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 20.62,
|
| 256 |
+
"grad_norm": 0.598238468170166,
|
| 257 |
+
"learning_rate": 2.5e-05,
|
| 258 |
+
"loss": 1.1566,
|
| 259 |
+
"step": 500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 20.99,
|
| 263 |
+
"eval_accuracy": 0.4405594405594406,
|
| 264 |
+
"eval_loss": 1.3691043853759766,
|
| 265 |
+
"eval_runtime": 6.6443,
|
| 266 |
+
"eval_samples_per_second": 43.044,
|
| 267 |
+
"eval_steps_per_second": 2.709,
|
| 268 |
+
"step": 509
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 21.98,
|
| 272 |
+
"eval_accuracy": 0.4825174825174825,
|
| 273 |
+
"eval_loss": 1.3120107650756836,
|
| 274 |
+
"eval_runtime": 4.9585,
|
| 275 |
+
"eval_samples_per_second": 57.679,
|
| 276 |
+
"eval_steps_per_second": 3.63,
|
| 277 |
+
"step": 533
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 22.68,
|
| 281 |
+
"grad_norm": 0.682925820350647,
|
| 282 |
+
"learning_rate": 2.75e-05,
|
| 283 |
+
"loss": 1.0676,
|
| 284 |
+
"step": 550
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 22.97,
|
| 288 |
+
"eval_accuracy": 0.486013986013986,
|
| 289 |
+
"eval_loss": 1.2839338779449463,
|
| 290 |
+
"eval_runtime": 4.0382,
|
| 291 |
+
"eval_samples_per_second": 70.824,
|
| 292 |
+
"eval_steps_per_second": 4.457,
|
| 293 |
+
"step": 557
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 24.0,
|
| 297 |
+
"eval_accuracy": 0.5104895104895105,
|
| 298 |
+
"eval_loss": 1.2549891471862793,
|
| 299 |
+
"eval_runtime": 5.1896,
|
| 300 |
+
"eval_samples_per_second": 55.11,
|
| 301 |
+
"eval_steps_per_second": 3.468,
|
| 302 |
+
"step": 582
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 24.74,
|
| 306 |
+
"grad_norm": 1.1368101835250854,
|
| 307 |
+
"learning_rate": 3e-05,
|
| 308 |
+
"loss": 0.992,
|
| 309 |
+
"step": 600
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 24.99,
|
| 313 |
+
"eval_accuracy": 0.5209790209790209,
|
| 314 |
+
"eval_loss": 1.2106566429138184,
|
| 315 |
+
"eval_runtime": 6.8941,
|
| 316 |
+
"eval_samples_per_second": 41.485,
|
| 317 |
+
"eval_steps_per_second": 2.611,
|
| 318 |
+
"step": 606
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 25.98,
|
| 322 |
+
"eval_accuracy": 0.5384615384615384,
|
| 323 |
+
"eval_loss": 1.1711338758468628,
|
| 324 |
+
"eval_runtime": 4.9707,
|
| 325 |
+
"eval_samples_per_second": 57.537,
|
| 326 |
+
"eval_steps_per_second": 3.621,
|
| 327 |
+
"step": 630
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 26.8,
|
| 331 |
+
"grad_norm": 0.9649831056594849,
|
| 332 |
+
"learning_rate": 2.9722222222222223e-05,
|
| 333 |
+
"loss": 0.9272,
|
| 334 |
+
"step": 650
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 26.97,
|
| 338 |
+
"eval_accuracy": 0.5594405594405595,
|
| 339 |
+
"eval_loss": 1.1318116188049316,
|
| 340 |
+
"eval_runtime": 5.5564,
|
| 341 |
+
"eval_samples_per_second": 51.472,
|
| 342 |
+
"eval_steps_per_second": 3.24,
|
| 343 |
+
"step": 654
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 28.0,
|
| 347 |
+
"eval_accuracy": 0.6153846153846154,
|
| 348 |
+
"eval_loss": 1.0594333410263062,
|
| 349 |
+
"eval_runtime": 4.6773,
|
| 350 |
+
"eval_samples_per_second": 61.147,
|
| 351 |
+
"eval_steps_per_second": 3.848,
|
| 352 |
+
"step": 679
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 28.87,
|
| 356 |
+
"grad_norm": 0.883937418460846,
|
| 357 |
+
"learning_rate": 2.9444444444444445e-05,
|
| 358 |
+
"loss": 0.8478,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 28.99,
|
| 363 |
+
"eval_accuracy": 0.6013986013986014,
|
| 364 |
+
"eval_loss": 1.054669737815857,
|
| 365 |
+
"eval_runtime": 4.9219,
|
| 366 |
+
"eval_samples_per_second": 58.108,
|
| 367 |
+
"eval_steps_per_second": 3.657,
|
| 368 |
+
"step": 703
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 29.98,
|
| 372 |
+
"eval_accuracy": 0.6363636363636364,
|
| 373 |
+
"eval_loss": 0.9822685122489929,
|
| 374 |
+
"eval_runtime": 6.3133,
|
| 375 |
+
"eval_samples_per_second": 45.302,
|
| 376 |
+
"eval_steps_per_second": 2.851,
|
| 377 |
+
"step": 727
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 30.93,
|
| 381 |
+
"grad_norm": 1.3742878437042236,
|
| 382 |
+
"learning_rate": 2.9166666666666666e-05,
|
| 383 |
+
"loss": 0.7627,
|
| 384 |
+
"step": 750
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 30.97,
|
| 388 |
+
"eval_accuracy": 0.6398601398601399,
|
| 389 |
+
"eval_loss": 1.00295090675354,
|
| 390 |
+
"eval_runtime": 6.154,
|
| 391 |
+
"eval_samples_per_second": 46.473,
|
| 392 |
+
"eval_steps_per_second": 2.925,
|
| 393 |
+
"step": 751
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 32.0,
|
| 397 |
+
"eval_accuracy": 0.6608391608391608,
|
| 398 |
+
"eval_loss": 0.930969774723053,
|
| 399 |
+
"eval_runtime": 5.6747,
|
| 400 |
+
"eval_samples_per_second": 50.399,
|
| 401 |
+
"eval_steps_per_second": 3.172,
|
| 402 |
+
"step": 776
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 32.99,
|
| 406 |
+
"grad_norm": 1.329268217086792,
|
| 407 |
+
"learning_rate": 2.8888888888888888e-05,
|
| 408 |
+
"loss": 0.7266,
|
| 409 |
+
"step": 800
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 32.99,
|
| 413 |
+
"eval_accuracy": 0.6678321678321678,
|
| 414 |
+
"eval_loss": 0.9228739738464355,
|
| 415 |
+
"eval_runtime": 5.382,
|
| 416 |
+
"eval_samples_per_second": 53.14,
|
| 417 |
+
"eval_steps_per_second": 3.344,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 33.98,
|
| 422 |
+
"eval_accuracy": 0.6958041958041958,
|
| 423 |
+
"eval_loss": 0.8684509992599487,
|
| 424 |
+
"eval_runtime": 4.8497,
|
| 425 |
+
"eval_samples_per_second": 58.973,
|
| 426 |
+
"eval_steps_per_second": 3.712,
|
| 427 |
+
"step": 824
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 34.97,
|
| 431 |
+
"eval_accuracy": 0.6643356643356644,
|
| 432 |
+
"eval_loss": 0.8954732418060303,
|
| 433 |
+
"eval_runtime": 5.2083,
|
| 434 |
+
"eval_samples_per_second": 54.912,
|
| 435 |
+
"eval_steps_per_second": 3.456,
|
| 436 |
+
"step": 848
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 35.05,
|
| 440 |
+
"grad_norm": 1.3892701864242554,
|
| 441 |
+
"learning_rate": 2.8611111111111113e-05,
|
| 442 |
+
"loss": 0.6906,
|
| 443 |
+
"step": 850
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 36.0,
|
| 447 |
+
"eval_accuracy": 0.6713286713286714,
|
| 448 |
+
"eval_loss": 0.9125654101371765,
|
| 449 |
+
"eval_runtime": 5.3068,
|
| 450 |
+
"eval_samples_per_second": 53.894,
|
| 451 |
+
"eval_steps_per_second": 3.392,
|
| 452 |
+
"step": 873
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 36.99,
|
| 456 |
+
"eval_accuracy": 0.6923076923076923,
|
| 457 |
+
"eval_loss": 0.8543534874916077,
|
| 458 |
+
"eval_runtime": 4.3351,
|
| 459 |
+
"eval_samples_per_second": 65.974,
|
| 460 |
+
"eval_steps_per_second": 4.152,
|
| 461 |
+
"step": 897
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 37.11,
|
| 465 |
+
"grad_norm": 0.836291491985321,
|
| 466 |
+
"learning_rate": 2.8333333333333332e-05,
|
| 467 |
+
"loss": 0.6721,
|
| 468 |
+
"step": 900
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 37.98,
|
| 472 |
+
"eval_accuracy": 0.6923076923076923,
|
| 473 |
+
"eval_loss": 0.8480322957038879,
|
| 474 |
+
"eval_runtime": 5.1861,
|
| 475 |
+
"eval_samples_per_second": 55.147,
|
| 476 |
+
"eval_steps_per_second": 3.471,
|
| 477 |
+
"step": 921
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 38.97,
|
| 481 |
+
"eval_accuracy": 0.7097902097902098,
|
| 482 |
+
"eval_loss": 0.8354606628417969,
|
| 483 |
+
"eval_runtime": 6.3247,
|
| 484 |
+
"eval_samples_per_second": 45.22,
|
| 485 |
+
"eval_steps_per_second": 2.846,
|
| 486 |
+
"step": 945
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 39.18,
|
| 490 |
+
"grad_norm": 1.6499431133270264,
|
| 491 |
+
"learning_rate": 2.8055555555555557e-05,
|
| 492 |
+
"loss": 0.6442,
|
| 493 |
+
"step": 950
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 40.0,
|
| 497 |
+
"eval_accuracy": 0.6958041958041958,
|
| 498 |
+
"eval_loss": 0.8412452340126038,
|
| 499 |
+
"eval_runtime": 5.2281,
|
| 500 |
+
"eval_samples_per_second": 54.704,
|
| 501 |
+
"eval_steps_per_second": 3.443,
|
| 502 |
+
"step": 970
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 40.99,
|
| 506 |
+
"eval_accuracy": 0.6888111888111889,
|
| 507 |
+
"eval_loss": 0.8356389999389648,
|
| 508 |
+
"eval_runtime": 4.8326,
|
| 509 |
+
"eval_samples_per_second": 59.181,
|
| 510 |
+
"eval_steps_per_second": 3.725,
|
| 511 |
+
"step": 994
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 41.24,
|
| 515 |
+
"grad_norm": 1.1766818761825562,
|
| 516 |
+
"learning_rate": 2.777777777777778e-05,
|
| 517 |
+
"loss": 0.6465,
|
| 518 |
+
"step": 1000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 41.98,
|
| 522 |
+
"eval_accuracy": 0.7062937062937062,
|
| 523 |
+
"eval_loss": 0.8180016875267029,
|
| 524 |
+
"eval_runtime": 5.7926,
|
| 525 |
+
"eval_samples_per_second": 49.374,
|
| 526 |
+
"eval_steps_per_second": 3.107,
|
| 527 |
+
"step": 1018
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 42.97,
|
| 531 |
+
"eval_accuracy": 0.7027972027972028,
|
| 532 |
+
"eval_loss": 0.8103991150856018,
|
| 533 |
+
"eval_runtime": 5.5185,
|
| 534 |
+
"eval_samples_per_second": 51.825,
|
| 535 |
+
"eval_steps_per_second": 3.262,
|
| 536 |
+
"step": 1042
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 43.3,
|
| 540 |
+
"grad_norm": 0.9722403287887573,
|
| 541 |
+
"learning_rate": 2.75e-05,
|
| 542 |
+
"loss": 0.6086,
|
| 543 |
+
"step": 1050
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 44.0,
|
| 547 |
+
"eval_accuracy": 0.6958041958041958,
|
| 548 |
+
"eval_loss": 0.8162235617637634,
|
| 549 |
+
"eval_runtime": 4.9174,
|
| 550 |
+
"eval_samples_per_second": 58.161,
|
| 551 |
+
"eval_steps_per_second": 3.66,
|
| 552 |
+
"step": 1067
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 44.99,
|
| 556 |
+
"eval_accuracy": 0.7027972027972028,
|
| 557 |
+
"eval_loss": 0.7957289218902588,
|
| 558 |
+
"eval_runtime": 4.6891,
|
| 559 |
+
"eval_samples_per_second": 60.992,
|
| 560 |
+
"eval_steps_per_second": 3.839,
|
| 561 |
+
"step": 1091
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 45.36,
|
| 565 |
+
"grad_norm": 1.269113302230835,
|
| 566 |
+
"learning_rate": 2.7222222222222223e-05,
|
| 567 |
+
"loss": 0.5863,
|
| 568 |
+
"step": 1100
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 45.98,
|
| 572 |
+
"eval_accuracy": 0.6958041958041958,
|
| 573 |
+
"eval_loss": 0.8143528699874878,
|
| 574 |
+
"eval_runtime": 6.6805,
|
| 575 |
+
"eval_samples_per_second": 42.811,
|
| 576 |
+
"eval_steps_per_second": 2.694,
|
| 577 |
+
"step": 1115
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 46.97,
|
| 581 |
+
"eval_accuracy": 0.7027972027972028,
|
| 582 |
+
"eval_loss": 0.78568434715271,
|
| 583 |
+
"eval_runtime": 4.7422,
|
| 584 |
+
"eval_samples_per_second": 60.31,
|
| 585 |
+
"eval_steps_per_second": 3.796,
|
| 586 |
+
"step": 1139
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 47.42,
|
| 590 |
+
"grad_norm": 0.9775255918502808,
|
| 591 |
+
"learning_rate": 2.6944444444444445e-05,
|
| 592 |
+
"loss": 0.5877,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 48.0,
|
| 597 |
+
"eval_accuracy": 0.7132867132867133,
|
| 598 |
+
"eval_loss": 0.7764595150947571,
|
| 599 |
+
"eval_runtime": 5.76,
|
| 600 |
+
"eval_samples_per_second": 49.653,
|
| 601 |
+
"eval_steps_per_second": 3.125,
|
| 602 |
+
"step": 1164
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 48.99,
|
| 606 |
+
"eval_accuracy": 0.6993006993006993,
|
| 607 |
+
"eval_loss": 0.7881478071212769,
|
| 608 |
+
"eval_runtime": 5.4965,
|
| 609 |
+
"eval_samples_per_second": 52.033,
|
| 610 |
+
"eval_steps_per_second": 3.275,
|
| 611 |
+
"step": 1188
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 49.48,
|
| 615 |
+
"grad_norm": 1.540124773979187,
|
| 616 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 617 |
+
"loss": 0.5629,
|
| 618 |
+
"step": 1200
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 49.98,
|
| 622 |
+
"eval_accuracy": 0.7097902097902098,
|
| 623 |
+
"eval_loss": 0.7658265829086304,
|
| 624 |
+
"eval_runtime": 4.731,
|
| 625 |
+
"eval_samples_per_second": 60.452,
|
| 626 |
+
"eval_steps_per_second": 3.805,
|
| 627 |
+
"step": 1212
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 50.97,
|
| 631 |
+
"eval_accuracy": 0.7132867132867133,
|
| 632 |
+
"eval_loss": 0.7723098397254944,
|
| 633 |
+
"eval_runtime": 5.8352,
|
| 634 |
+
"eval_samples_per_second": 49.013,
|
| 635 |
+
"eval_steps_per_second": 3.085,
|
| 636 |
+
"step": 1236
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 51.55,
|
| 640 |
+
"grad_norm": 1.2498500347137451,
|
| 641 |
+
"learning_rate": 2.6388888888888892e-05,
|
| 642 |
+
"loss": 0.5476,
|
| 643 |
+
"step": 1250
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 52.0,
|
| 647 |
+
"eval_accuracy": 0.7097902097902098,
|
| 648 |
+
"eval_loss": 0.7603952884674072,
|
| 649 |
+
"eval_runtime": 4.448,
|
| 650 |
+
"eval_samples_per_second": 64.299,
|
| 651 |
+
"eval_steps_per_second": 4.047,
|
| 652 |
+
"step": 1261
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 52.99,
|
| 656 |
+
"eval_accuracy": 0.7202797202797203,
|
| 657 |
+
"eval_loss": 0.7554137706756592,
|
| 658 |
+
"eval_runtime": 6.4218,
|
| 659 |
+
"eval_samples_per_second": 44.536,
|
| 660 |
+
"eval_steps_per_second": 2.803,
|
| 661 |
+
"step": 1285
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 53.61,
|
| 665 |
+
"grad_norm": 0.9919388890266418,
|
| 666 |
+
"learning_rate": 2.6116666666666667e-05,
|
| 667 |
+
"loss": 0.5357,
|
| 668 |
+
"step": 1300
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 53.98,
|
| 672 |
+
"eval_accuracy": 0.7307692307692307,
|
| 673 |
+
"eval_loss": 0.7458928227424622,
|
| 674 |
+
"eval_runtime": 5.3791,
|
| 675 |
+
"eval_samples_per_second": 53.168,
|
| 676 |
+
"eval_steps_per_second": 3.346,
|
| 677 |
+
"step": 1309
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 54.97,
|
| 681 |
+
"eval_accuracy": 0.7132867132867133,
|
| 682 |
+
"eval_loss": 0.7632877230644226,
|
| 683 |
+
"eval_runtime": 5.278,
|
| 684 |
+
"eval_samples_per_second": 54.187,
|
| 685 |
+
"eval_steps_per_second": 3.41,
|
| 686 |
+
"step": 1333
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 55.67,
|
| 690 |
+
"grad_norm": 1.688183307647705,
|
| 691 |
+
"learning_rate": 2.5838888888888892e-05,
|
| 692 |
+
"loss": 0.5335,
|
| 693 |
+
"step": 1350
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 56.0,
|
| 697 |
+
"eval_accuracy": 0.7167832167832168,
|
| 698 |
+
"eval_loss": 0.768308162689209,
|
| 699 |
+
"eval_runtime": 5.7022,
|
| 700 |
+
"eval_samples_per_second": 50.156,
|
| 701 |
+
"eval_steps_per_second": 3.157,
|
| 702 |
+
"step": 1358
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 56.99,
|
| 706 |
+
"eval_accuracy": 0.7307692307692307,
|
| 707 |
+
"eval_loss": 0.7380541563034058,
|
| 708 |
+
"eval_runtime": 4.522,
|
| 709 |
+
"eval_samples_per_second": 63.247,
|
| 710 |
+
"eval_steps_per_second": 3.981,
|
| 711 |
+
"step": 1382
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 57.73,
|
| 715 |
+
"grad_norm": 1.4895784854888916,
|
| 716 |
+
"learning_rate": 2.556111111111111e-05,
|
| 717 |
+
"loss": 0.5107,
|
| 718 |
+
"step": 1400
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 57.98,
|
| 722 |
+
"eval_accuracy": 0.7377622377622378,
|
| 723 |
+
"eval_loss": 0.7308338284492493,
|
| 724 |
+
"eval_runtime": 4.4787,
|
| 725 |
+
"eval_samples_per_second": 63.857,
|
| 726 |
+
"eval_steps_per_second": 4.019,
|
| 727 |
+
"step": 1406
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 58.97,
|
| 731 |
+
"eval_accuracy": 0.7237762237762237,
|
| 732 |
+
"eval_loss": 0.7441032528877258,
|
| 733 |
+
"eval_runtime": 5.8744,
|
| 734 |
+
"eval_samples_per_second": 48.685,
|
| 735 |
+
"eval_steps_per_second": 3.064,
|
| 736 |
+
"step": 1430
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 59.79,
|
| 740 |
+
"grad_norm": 1.4925004243850708,
|
| 741 |
+
"learning_rate": 2.5283333333333332e-05,
|
| 742 |
+
"loss": 0.5105,
|
| 743 |
+
"step": 1450
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 60.0,
|
| 747 |
+
"eval_accuracy": 0.7307692307692307,
|
| 748 |
+
"eval_loss": 0.7481815218925476,
|
| 749 |
+
"eval_runtime": 7.272,
|
| 750 |
+
"eval_samples_per_second": 39.329,
|
| 751 |
+
"eval_steps_per_second": 2.475,
|
| 752 |
+
"step": 1455
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 60.99,
|
| 756 |
+
"eval_accuracy": 0.7342657342657343,
|
| 757 |
+
"eval_loss": 0.733482301235199,
|
| 758 |
+
"eval_runtime": 4.6235,
|
| 759 |
+
"eval_samples_per_second": 61.858,
|
| 760 |
+
"eval_steps_per_second": 3.893,
|
| 761 |
+
"step": 1479
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 61.86,
|
| 765 |
+
"grad_norm": 1.3200663328170776,
|
| 766 |
+
"learning_rate": 2.5005555555555558e-05,
|
| 767 |
+
"loss": 0.4914,
|
| 768 |
+
"step": 1500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 61.98,
|
| 772 |
+
"eval_accuracy": 0.7447552447552448,
|
| 773 |
+
"eval_loss": 0.7241908311843872,
|
| 774 |
+
"eval_runtime": 4.8198,
|
| 775 |
+
"eval_samples_per_second": 59.338,
|
| 776 |
+
"eval_steps_per_second": 3.735,
|
| 777 |
+
"step": 1503
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 62.97,
|
| 781 |
+
"eval_accuracy": 0.7377622377622378,
|
| 782 |
+
"eval_loss": 0.7321043014526367,
|
| 783 |
+
"eval_runtime": 5.8929,
|
| 784 |
+
"eval_samples_per_second": 48.533,
|
| 785 |
+
"eval_steps_per_second": 3.055,
|
| 786 |
+
"step": 1527
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 63.92,
|
| 790 |
+
"grad_norm": 1.1309747695922852,
|
| 791 |
+
"learning_rate": 2.472777777777778e-05,
|
| 792 |
+
"loss": 0.4839,
|
| 793 |
+
"step": 1550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 64.0,
|
| 797 |
+
"eval_accuracy": 0.7342657342657343,
|
| 798 |
+
"eval_loss": 0.7220665216445923,
|
| 799 |
+
"eval_runtime": 5.8635,
|
| 800 |
+
"eval_samples_per_second": 48.776,
|
| 801 |
+
"eval_steps_per_second": 3.07,
|
| 802 |
+
"step": 1552
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 64.99,
|
| 806 |
+
"eval_accuracy": 0.7412587412587412,
|
| 807 |
+
"eval_loss": 0.7136482000350952,
|
| 808 |
+
"eval_runtime": 4.3102,
|
| 809 |
+
"eval_samples_per_second": 66.354,
|
| 810 |
+
"eval_steps_per_second": 4.176,
|
| 811 |
+
"step": 1576
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 65.98,
|
| 815 |
+
"grad_norm": 1.1314157247543335,
|
| 816 |
+
"learning_rate": 2.4449999999999998e-05,
|
| 817 |
+
"loss": 0.4751,
|
| 818 |
+
"step": 1600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 65.98,
|
| 822 |
+
"eval_accuracy": 0.7412587412587412,
|
| 823 |
+
"eval_loss": 0.7198111414909363,
|
| 824 |
+
"eval_runtime": 4.7841,
|
| 825 |
+
"eval_samples_per_second": 59.781,
|
| 826 |
+
"eval_steps_per_second": 3.762,
|
| 827 |
+
"step": 1600
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 66.97,
|
| 831 |
+
"eval_accuracy": 0.7377622377622378,
|
| 832 |
+
"eval_loss": 0.7145721912384033,
|
| 833 |
+
"eval_runtime": 6.347,
|
| 834 |
+
"eval_samples_per_second": 45.061,
|
| 835 |
+
"eval_steps_per_second": 2.836,
|
| 836 |
+
"step": 1624
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 68.0,
|
| 840 |
+
"eval_accuracy": 0.7447552447552448,
|
| 841 |
+
"eval_loss": 0.6970916390419006,
|
| 842 |
+
"eval_runtime": 5.6871,
|
| 843 |
+
"eval_samples_per_second": 50.289,
|
| 844 |
+
"eval_steps_per_second": 3.165,
|
| 845 |
+
"step": 1649
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 68.04,
|
| 849 |
+
"grad_norm": 2.397585153579712,
|
| 850 |
+
"learning_rate": 2.4172222222222223e-05,
|
| 851 |
+
"loss": 0.4639,
|
| 852 |
+
"step": 1650
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 68.99,
|
| 856 |
+
"eval_accuracy": 0.7272727272727273,
|
| 857 |
+
"eval_loss": 0.7201464176177979,
|
| 858 |
+
"eval_runtime": 4.4157,
|
| 859 |
+
"eval_samples_per_second": 64.769,
|
| 860 |
+
"eval_steps_per_second": 4.076,
|
| 861 |
+
"step": 1673
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 69.98,
|
| 865 |
+
"eval_accuracy": 0.7307692307692307,
|
| 866 |
+
"eval_loss": 0.7244682312011719,
|
| 867 |
+
"eval_runtime": 5.4392,
|
| 868 |
+
"eval_samples_per_second": 52.581,
|
| 869 |
+
"eval_steps_per_second": 3.309,
|
| 870 |
+
"step": 1697
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 70.1,
|
| 874 |
+
"grad_norm": 2.062610387802124,
|
| 875 |
+
"learning_rate": 2.3894444444444445e-05,
|
| 876 |
+
"loss": 0.4581,
|
| 877 |
+
"step": 1700
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 70.97,
|
| 881 |
+
"eval_accuracy": 0.7447552447552448,
|
| 882 |
+
"eval_loss": 0.7077587842941284,
|
| 883 |
+
"eval_runtime": 5.1002,
|
| 884 |
+
"eval_samples_per_second": 56.076,
|
| 885 |
+
"eval_steps_per_second": 3.529,
|
| 886 |
+
"step": 1721
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 72.0,
|
| 890 |
+
"eval_accuracy": 0.7517482517482518,
|
| 891 |
+
"eval_loss": 0.6957913637161255,
|
| 892 |
+
"eval_runtime": 4.4485,
|
| 893 |
+
"eval_samples_per_second": 64.291,
|
| 894 |
+
"eval_steps_per_second": 4.046,
|
| 895 |
+
"step": 1746
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 72.16,
|
| 899 |
+
"grad_norm": 2.7808456420898438,
|
| 900 |
+
"learning_rate": 2.3616666666666667e-05,
|
| 901 |
+
"loss": 0.4643,
|
| 902 |
+
"step": 1750
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 72.99,
|
| 906 |
+
"eval_accuracy": 0.7447552447552448,
|
| 907 |
+
"eval_loss": 0.7036928534507751,
|
| 908 |
+
"eval_runtime": 5.9101,
|
| 909 |
+
"eval_samples_per_second": 48.392,
|
| 910 |
+
"eval_steps_per_second": 3.046,
|
| 911 |
+
"step": 1770
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 73.98,
|
| 915 |
+
"eval_accuracy": 0.7482517482517482,
|
| 916 |
+
"eval_loss": 0.71629399061203,
|
| 917 |
+
"eval_runtime": 6.0211,
|
| 918 |
+
"eval_samples_per_second": 47.5,
|
| 919 |
+
"eval_steps_per_second": 2.989,
|
| 920 |
+
"step": 1794
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 74.23,
|
| 924 |
+
"grad_norm": 1.78495192527771,
|
| 925 |
+
"learning_rate": 2.333888888888889e-05,
|
| 926 |
+
"loss": 0.442,
|
| 927 |
+
"step": 1800
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 74.97,
|
| 931 |
+
"eval_accuracy": 0.7377622377622378,
|
| 932 |
+
"eval_loss": 0.6997957229614258,
|
| 933 |
+
"eval_runtime": 4.4212,
|
| 934 |
+
"eval_samples_per_second": 64.688,
|
| 935 |
+
"eval_steps_per_second": 4.071,
|
| 936 |
+
"step": 1818
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 76.0,
|
| 940 |
+
"eval_accuracy": 0.7447552447552448,
|
| 941 |
+
"eval_loss": 0.6946483850479126,
|
| 942 |
+
"eval_runtime": 4.0507,
|
| 943 |
+
"eval_samples_per_second": 70.605,
|
| 944 |
+
"eval_steps_per_second": 4.444,
|
| 945 |
+
"step": 1843
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 76.29,
|
| 949 |
+
"grad_norm": 1.7383118867874146,
|
| 950 |
+
"learning_rate": 2.306111111111111e-05,
|
| 951 |
+
"loss": 0.4305,
|
| 952 |
+
"step": 1850
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 76.99,
|
| 956 |
+
"eval_accuracy": 0.7552447552447552,
|
| 957 |
+
"eval_loss": 0.6857091784477234,
|
| 958 |
+
"eval_runtime": 4.1718,
|
| 959 |
+
"eval_samples_per_second": 68.556,
|
| 960 |
+
"eval_steps_per_second": 4.315,
|
| 961 |
+
"step": 1867
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 77.98,
|
| 965 |
+
"eval_accuracy": 0.7447552447552448,
|
| 966 |
+
"eval_loss": 0.6936307549476624,
|
| 967 |
+
"eval_runtime": 3.8781,
|
| 968 |
+
"eval_samples_per_second": 73.747,
|
| 969 |
+
"eval_steps_per_second": 4.641,
|
| 970 |
+
"step": 1891
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 78.35,
|
| 974 |
+
"grad_norm": 1.047067403793335,
|
| 975 |
+
"learning_rate": 2.2783333333333336e-05,
|
| 976 |
+
"loss": 0.4416,
|
| 977 |
+
"step": 1900
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 78.97,
|
| 981 |
+
"eval_accuracy": 0.7517482517482518,
|
| 982 |
+
"eval_loss": 0.6965110301971436,
|
| 983 |
+
"eval_runtime": 5.1318,
|
| 984 |
+
"eval_samples_per_second": 55.731,
|
| 985 |
+
"eval_steps_per_second": 3.508,
|
| 986 |
+
"step": 1915
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 80.0,
|
| 990 |
+
"eval_accuracy": 0.7482517482517482,
|
| 991 |
+
"eval_loss": 0.7017127871513367,
|
| 992 |
+
"eval_runtime": 4.3418,
|
| 993 |
+
"eval_samples_per_second": 65.871,
|
| 994 |
+
"eval_steps_per_second": 4.146,
|
| 995 |
+
"step": 1940
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 80.41,
|
| 999 |
+
"grad_norm": 1.5354928970336914,
|
| 1000 |
+
"learning_rate": 2.2505555555555554e-05,
|
| 1001 |
+
"loss": 0.428,
|
| 1002 |
+
"step": 1950
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 80.99,
|
| 1006 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1007 |
+
"eval_loss": 0.6970596313476562,
|
| 1008 |
+
"eval_runtime": 5.973,
|
| 1009 |
+
"eval_samples_per_second": 47.882,
|
| 1010 |
+
"eval_steps_per_second": 3.014,
|
| 1011 |
+
"step": 1964
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 81.98,
|
| 1015 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1016 |
+
"eval_loss": 0.6897542476654053,
|
| 1017 |
+
"eval_runtime": 5.0481,
|
| 1018 |
+
"eval_samples_per_second": 56.655,
|
| 1019 |
+
"eval_steps_per_second": 3.566,
|
| 1020 |
+
"step": 1988
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 82.47,
|
| 1024 |
+
"grad_norm": 1.7141317129135132,
|
| 1025 |
+
"learning_rate": 2.2227777777777776e-05,
|
| 1026 |
+
"loss": 0.4093,
|
| 1027 |
+
"step": 2000
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 82.97,
|
| 1031 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1032 |
+
"eval_loss": 0.7004020810127258,
|
| 1033 |
+
"eval_runtime": 4.1986,
|
| 1034 |
+
"eval_samples_per_second": 68.118,
|
| 1035 |
+
"eval_steps_per_second": 4.287,
|
| 1036 |
+
"step": 2012
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 84.0,
|
| 1040 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1041 |
+
"eval_loss": 0.6867479681968689,
|
| 1042 |
+
"eval_runtime": 4.6871,
|
| 1043 |
+
"eval_samples_per_second": 61.018,
|
| 1044 |
+
"eval_steps_per_second": 3.84,
|
| 1045 |
+
"step": 2037
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 84.54,
|
| 1049 |
+
"grad_norm": 2.0219666957855225,
|
| 1050 |
+
"learning_rate": 2.195e-05,
|
| 1051 |
+
"loss": 0.4148,
|
| 1052 |
+
"step": 2050
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 84.99,
|
| 1056 |
+
"eval_accuracy": 0.7377622377622378,
|
| 1057 |
+
"eval_loss": 0.7070020437240601,
|
| 1058 |
+
"eval_runtime": 5.9326,
|
| 1059 |
+
"eval_samples_per_second": 48.208,
|
| 1060 |
+
"eval_steps_per_second": 3.034,
|
| 1061 |
+
"step": 2061
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 85.98,
|
| 1065 |
+
"eval_accuracy": 0.7447552447552448,
|
| 1066 |
+
"eval_loss": 0.7030305862426758,
|
| 1067 |
+
"eval_runtime": 5.3564,
|
| 1068 |
+
"eval_samples_per_second": 53.394,
|
| 1069 |
+
"eval_steps_per_second": 3.36,
|
| 1070 |
+
"step": 2085
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 86.6,
|
| 1074 |
+
"grad_norm": 1.4678714275360107,
|
| 1075 |
+
"learning_rate": 2.1672222222222223e-05,
|
| 1076 |
+
"loss": 0.3923,
|
| 1077 |
+
"step": 2100
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 86.97,
|
| 1081 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1082 |
+
"eval_loss": 0.678174614906311,
|
| 1083 |
+
"eval_runtime": 3.9745,
|
| 1084 |
+
"eval_samples_per_second": 71.96,
|
| 1085 |
+
"eval_steps_per_second": 4.529,
|
| 1086 |
+
"step": 2109
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 88.0,
|
| 1090 |
+
"eval_accuracy": 0.7412587412587412,
|
| 1091 |
+
"eval_loss": 0.7166118621826172,
|
| 1092 |
+
"eval_runtime": 4.0358,
|
| 1093 |
+
"eval_samples_per_second": 70.866,
|
| 1094 |
+
"eval_steps_per_second": 4.46,
|
| 1095 |
+
"step": 2134
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 88.66,
|
| 1099 |
+
"grad_norm": 1.589543342590332,
|
| 1100 |
+
"learning_rate": 2.1394444444444445e-05,
|
| 1101 |
+
"loss": 0.3964,
|
| 1102 |
+
"step": 2150
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 88.99,
|
| 1106 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1107 |
+
"eval_loss": 0.7075912952423096,
|
| 1108 |
+
"eval_runtime": 5.0331,
|
| 1109 |
+
"eval_samples_per_second": 56.823,
|
| 1110 |
+
"eval_steps_per_second": 3.576,
|
| 1111 |
+
"step": 2158
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 89.98,
|
| 1115 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1116 |
+
"eval_loss": 0.6867172122001648,
|
| 1117 |
+
"eval_runtime": 5.386,
|
| 1118 |
+
"eval_samples_per_second": 53.101,
|
| 1119 |
+
"eval_steps_per_second": 3.342,
|
| 1120 |
+
"step": 2182
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 90.72,
|
| 1124 |
+
"grad_norm": 1.3886605501174927,
|
| 1125 |
+
"learning_rate": 2.1116666666666667e-05,
|
| 1126 |
+
"loss": 0.3846,
|
| 1127 |
+
"step": 2200
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 90.97,
|
| 1131 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1132 |
+
"eval_loss": 0.6913285851478577,
|
| 1133 |
+
"eval_runtime": 5.5324,
|
| 1134 |
+
"eval_samples_per_second": 51.696,
|
| 1135 |
+
"eval_steps_per_second": 3.254,
|
| 1136 |
+
"step": 2206
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 92.0,
|
| 1140 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1141 |
+
"eval_loss": 0.7160294651985168,
|
| 1142 |
+
"eval_runtime": 5.2753,
|
| 1143 |
+
"eval_samples_per_second": 54.215,
|
| 1144 |
+
"eval_steps_per_second": 3.412,
|
| 1145 |
+
"step": 2231
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 92.78,
|
| 1149 |
+
"grad_norm": 2.4106783866882324,
|
| 1150 |
+
"learning_rate": 2.083888888888889e-05,
|
| 1151 |
+
"loss": 0.3654,
|
| 1152 |
+
"step": 2250
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 92.99,
|
| 1156 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1157 |
+
"eval_loss": 0.6765207052230835,
|
| 1158 |
+
"eval_runtime": 5.5671,
|
| 1159 |
+
"eval_samples_per_second": 51.373,
|
| 1160 |
+
"eval_steps_per_second": 3.233,
|
| 1161 |
+
"step": 2255
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 93.98,
|
| 1165 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1166 |
+
"eval_loss": 0.6881967186927795,
|
| 1167 |
+
"eval_runtime": 3.8228,
|
| 1168 |
+
"eval_samples_per_second": 74.814,
|
| 1169 |
+
"eval_steps_per_second": 4.709,
|
| 1170 |
+
"step": 2279
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 94.85,
|
| 1174 |
+
"grad_norm": 0.8871183395385742,
|
| 1175 |
+
"learning_rate": 2.0561111111111114e-05,
|
| 1176 |
+
"loss": 0.3577,
|
| 1177 |
+
"step": 2300
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 94.97,
|
| 1181 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1182 |
+
"eval_loss": 0.6852585673332214,
|
| 1183 |
+
"eval_runtime": 4.7228,
|
| 1184 |
+
"eval_samples_per_second": 60.557,
|
| 1185 |
+
"eval_steps_per_second": 3.811,
|
| 1186 |
+
"step": 2303
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 96.0,
|
| 1190 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1191 |
+
"eval_loss": 0.7158808708190918,
|
| 1192 |
+
"eval_runtime": 5.6504,
|
| 1193 |
+
"eval_samples_per_second": 50.616,
|
| 1194 |
+
"eval_steps_per_second": 3.186,
|
| 1195 |
+
"step": 2328
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 96.91,
|
| 1199 |
+
"grad_norm": 1.0019863843917847,
|
| 1200 |
+
"learning_rate": 2.0283333333333333e-05,
|
| 1201 |
+
"loss": 0.37,
|
| 1202 |
+
"step": 2350
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 96.99,
|
| 1206 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1207 |
+
"eval_loss": 0.6943120360374451,
|
| 1208 |
+
"eval_runtime": 4.8337,
|
| 1209 |
+
"eval_samples_per_second": 59.168,
|
| 1210 |
+
"eval_steps_per_second": 3.724,
|
| 1211 |
+
"step": 2352
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 97.98,
|
| 1215 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1216 |
+
"eval_loss": 0.7010317444801331,
|
| 1217 |
+
"eval_runtime": 4.6874,
|
| 1218 |
+
"eval_samples_per_second": 61.015,
|
| 1219 |
+
"eval_steps_per_second": 3.84,
|
| 1220 |
+
"step": 2376
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 98.97,
|
| 1224 |
+
"grad_norm": 1.2908928394317627,
|
| 1225 |
+
"learning_rate": 2.0005555555555555e-05,
|
| 1226 |
+
"loss": 0.3473,
|
| 1227 |
+
"step": 2400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 98.97,
|
| 1231 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1232 |
+
"eval_loss": 0.693758487701416,
|
| 1233 |
+
"eval_runtime": 4.7585,
|
| 1234 |
+
"eval_samples_per_second": 60.103,
|
| 1235 |
+
"eval_steps_per_second": 3.783,
|
| 1236 |
+
"step": 2400
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 100.0,
|
| 1240 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1241 |
+
"eval_loss": 0.6918778419494629,
|
| 1242 |
+
"eval_runtime": 6.6891,
|
| 1243 |
+
"eval_samples_per_second": 42.756,
|
| 1244 |
+
"eval_steps_per_second": 2.691,
|
| 1245 |
+
"step": 2425
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 100.99,
|
| 1249 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1250 |
+
"eval_loss": 0.6849302053451538,
|
| 1251 |
+
"eval_runtime": 4.4685,
|
| 1252 |
+
"eval_samples_per_second": 64.003,
|
| 1253 |
+
"eval_steps_per_second": 4.028,
|
| 1254 |
+
"step": 2449
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 101.03,
|
| 1258 |
+
"grad_norm": 1.1730871200561523,
|
| 1259 |
+
"learning_rate": 1.972777777777778e-05,
|
| 1260 |
+
"loss": 0.3587,
|
| 1261 |
+
"step": 2450
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 101.98,
|
| 1265 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1266 |
+
"eval_loss": 0.6855939030647278,
|
| 1267 |
+
"eval_runtime": 4.3434,
|
| 1268 |
+
"eval_samples_per_second": 65.847,
|
| 1269 |
+
"eval_steps_per_second": 4.144,
|
| 1270 |
+
"step": 2473
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 102.97,
|
| 1274 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1275 |
+
"eval_loss": 0.7046144604682922,
|
| 1276 |
+
"eval_runtime": 4.7166,
|
| 1277 |
+
"eval_samples_per_second": 60.637,
|
| 1278 |
+
"eval_steps_per_second": 3.816,
|
| 1279 |
+
"step": 2497
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 103.09,
|
| 1283 |
+
"grad_norm": 1.3693217039108276,
|
| 1284 |
+
"learning_rate": 1.945e-05,
|
| 1285 |
+
"loss": 0.3429,
|
| 1286 |
+
"step": 2500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 104.0,
|
| 1290 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1291 |
+
"eval_loss": 0.6892997622489929,
|
| 1292 |
+
"eval_runtime": 5.3868,
|
| 1293 |
+
"eval_samples_per_second": 53.092,
|
| 1294 |
+
"eval_steps_per_second": 3.341,
|
| 1295 |
+
"step": 2522
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 104.99,
|
| 1299 |
+
"eval_accuracy": 0.7622377622377622,
|
| 1300 |
+
"eval_loss": 0.6913393139839172,
|
| 1301 |
+
"eval_runtime": 5.09,
|
| 1302 |
+
"eval_samples_per_second": 56.188,
|
| 1303 |
+
"eval_steps_per_second": 3.536,
|
| 1304 |
+
"step": 2546
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 105.15,
|
| 1308 |
+
"grad_norm": 1.923829436302185,
|
| 1309 |
+
"learning_rate": 1.9172222222222224e-05,
|
| 1310 |
+
"loss": 0.3549,
|
| 1311 |
+
"step": 2550
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 105.98,
|
| 1315 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1316 |
+
"eval_loss": 0.6880810856819153,
|
| 1317 |
+
"eval_runtime": 4.6668,
|
| 1318 |
+
"eval_samples_per_second": 61.283,
|
| 1319 |
+
"eval_steps_per_second": 3.857,
|
| 1320 |
+
"step": 2570
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 106.97,
|
| 1324 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1325 |
+
"eval_loss": 0.7097887396812439,
|
| 1326 |
+
"eval_runtime": 6.4652,
|
| 1327 |
+
"eval_samples_per_second": 44.237,
|
| 1328 |
+
"eval_steps_per_second": 2.784,
|
| 1329 |
+
"step": 2594
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 107.22,
|
| 1333 |
+
"grad_norm": 2.702012062072754,
|
| 1334 |
+
"learning_rate": 1.8894444444444446e-05,
|
| 1335 |
+
"loss": 0.3403,
|
| 1336 |
+
"step": 2600
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 108.0,
|
| 1340 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1341 |
+
"eval_loss": 0.6878336668014526,
|
| 1342 |
+
"eval_runtime": 4.6923,
|
| 1343 |
+
"eval_samples_per_second": 60.951,
|
| 1344 |
+
"eval_steps_per_second": 3.836,
|
| 1345 |
+
"step": 2619
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 108.99,
|
| 1349 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1350 |
+
"eval_loss": 0.695954442024231,
|
| 1351 |
+
"eval_runtime": 4.4809,
|
| 1352 |
+
"eval_samples_per_second": 63.827,
|
| 1353 |
+
"eval_steps_per_second": 4.017,
|
| 1354 |
+
"step": 2643
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 109.28,
|
| 1358 |
+
"grad_norm": 2.3427536487579346,
|
| 1359 |
+
"learning_rate": 1.8616666666666667e-05,
|
| 1360 |
+
"loss": 0.3253,
|
| 1361 |
+
"step": 2650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 109.98,
|
| 1365 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1366 |
+
"eval_loss": 0.7005948424339294,
|
| 1367 |
+
"eval_runtime": 4.8882,
|
| 1368 |
+
"eval_samples_per_second": 58.508,
|
| 1369 |
+
"eval_steps_per_second": 3.682,
|
| 1370 |
+
"step": 2667
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 110.97,
|
| 1374 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1375 |
+
"eval_loss": 0.6916196346282959,
|
| 1376 |
+
"eval_runtime": 5.2891,
|
| 1377 |
+
"eval_samples_per_second": 54.073,
|
| 1378 |
+
"eval_steps_per_second": 3.403,
|
| 1379 |
+
"step": 2691
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 111.34,
|
| 1383 |
+
"grad_norm": 2.178089141845703,
|
| 1384 |
+
"learning_rate": 1.833888888888889e-05,
|
| 1385 |
+
"loss": 0.3332,
|
| 1386 |
+
"step": 2700
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 112.0,
|
| 1390 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1391 |
+
"eval_loss": 0.7059447765350342,
|
| 1392 |
+
"eval_runtime": 4.7437,
|
| 1393 |
+
"eval_samples_per_second": 60.291,
|
| 1394 |
+
"eval_steps_per_second": 3.795,
|
| 1395 |
+
"step": 2716
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 112.99,
|
| 1399 |
+
"eval_accuracy": 0.7867132867132867,
|
| 1400 |
+
"eval_loss": 0.6904045939445496,
|
| 1401 |
+
"eval_runtime": 4.9942,
|
| 1402 |
+
"eval_samples_per_second": 57.267,
|
| 1403 |
+
"eval_steps_per_second": 3.604,
|
| 1404 |
+
"step": 2740
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 113.4,
|
| 1408 |
+
"grad_norm": 1.1625444889068604,
|
| 1409 |
+
"learning_rate": 1.806111111111111e-05,
|
| 1410 |
+
"loss": 0.3188,
|
| 1411 |
+
"step": 2750
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 113.98,
|
| 1415 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1416 |
+
"eval_loss": 0.6970774531364441,
|
| 1417 |
+
"eval_runtime": 6.4809,
|
| 1418 |
+
"eval_samples_per_second": 44.13,
|
| 1419 |
+
"eval_steps_per_second": 2.777,
|
| 1420 |
+
"step": 2764
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 114.97,
|
| 1424 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1425 |
+
"eval_loss": 0.700820803642273,
|
| 1426 |
+
"eval_runtime": 5.2617,
|
| 1427 |
+
"eval_samples_per_second": 54.355,
|
| 1428 |
+
"eval_steps_per_second": 3.421,
|
| 1429 |
+
"step": 2788
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 115.46,
|
| 1433 |
+
"grad_norm": 1.2394715547561646,
|
| 1434 |
+
"learning_rate": 1.7783333333333333e-05,
|
| 1435 |
+
"loss": 0.3112,
|
| 1436 |
+
"step": 2800
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 116.0,
|
| 1440 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1441 |
+
"eval_loss": 0.7002130150794983,
|
| 1442 |
+
"eval_runtime": 5.0937,
|
| 1443 |
+
"eval_samples_per_second": 56.147,
|
| 1444 |
+
"eval_steps_per_second": 3.534,
|
| 1445 |
+
"step": 2813
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 116.99,
|
| 1449 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1450 |
+
"eval_loss": 0.6909505724906921,
|
| 1451 |
+
"eval_runtime": 4.7575,
|
| 1452 |
+
"eval_samples_per_second": 60.116,
|
| 1453 |
+
"eval_steps_per_second": 3.784,
|
| 1454 |
+
"step": 2837
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 117.53,
|
| 1458 |
+
"grad_norm": 2.4334964752197266,
|
| 1459 |
+
"learning_rate": 1.7505555555555558e-05,
|
| 1460 |
+
"loss": 0.3153,
|
| 1461 |
+
"step": 2850
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 117.98,
|
| 1465 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1466 |
+
"eval_loss": 0.6957750916481018,
|
| 1467 |
+
"eval_runtime": 4.8105,
|
| 1468 |
+
"eval_samples_per_second": 59.453,
|
| 1469 |
+
"eval_steps_per_second": 3.742,
|
| 1470 |
+
"step": 2861
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 118.97,
|
| 1474 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1475 |
+
"eval_loss": 0.6867520213127136,
|
| 1476 |
+
"eval_runtime": 4.5411,
|
| 1477 |
+
"eval_samples_per_second": 62.98,
|
| 1478 |
+
"eval_steps_per_second": 3.964,
|
| 1479 |
+
"step": 2885
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 119.59,
|
| 1483 |
+
"grad_norm": 0.769097089767456,
|
| 1484 |
+
"learning_rate": 1.7227777777777777e-05,
|
| 1485 |
+
"loss": 0.3006,
|
| 1486 |
+
"step": 2900
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 120.0,
|
| 1490 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1491 |
+
"eval_loss": 0.6890790462493896,
|
| 1492 |
+
"eval_runtime": 4.5864,
|
| 1493 |
+
"eval_samples_per_second": 62.358,
|
| 1494 |
+
"eval_steps_per_second": 3.925,
|
| 1495 |
+
"step": 2910
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 120.99,
|
| 1499 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1500 |
+
"eval_loss": 0.6889089941978455,
|
| 1501 |
+
"eval_runtime": 6.5804,
|
| 1502 |
+
"eval_samples_per_second": 43.462,
|
| 1503 |
+
"eval_steps_per_second": 2.735,
|
| 1504 |
+
"step": 2934
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 121.65,
|
| 1508 |
+
"grad_norm": 1.8714542388916016,
|
| 1509 |
+
"learning_rate": 1.695e-05,
|
| 1510 |
+
"loss": 0.2967,
|
| 1511 |
+
"step": 2950
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 121.98,
|
| 1515 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1516 |
+
"eval_loss": 0.6935350894927979,
|
| 1517 |
+
"eval_runtime": 4.7491,
|
| 1518 |
+
"eval_samples_per_second": 60.223,
|
| 1519 |
+
"eval_steps_per_second": 3.79,
|
| 1520 |
+
"step": 2958
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 122.97,
|
| 1524 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1525 |
+
"eval_loss": 0.7058219909667969,
|
| 1526 |
+
"eval_runtime": 4.8941,
|
| 1527 |
+
"eval_samples_per_second": 58.438,
|
| 1528 |
+
"eval_steps_per_second": 3.678,
|
| 1529 |
+
"step": 2982
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 123.71,
|
| 1533 |
+
"grad_norm": 2.062924385070801,
|
| 1534 |
+
"learning_rate": 1.6672222222222224e-05,
|
| 1535 |
+
"loss": 0.2939,
|
| 1536 |
+
"step": 3000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 124.0,
|
| 1540 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1541 |
+
"eval_loss": 0.7220865488052368,
|
| 1542 |
+
"eval_runtime": 5.0487,
|
| 1543 |
+
"eval_samples_per_second": 56.648,
|
| 1544 |
+
"eval_steps_per_second": 3.565,
|
| 1545 |
+
"step": 3007
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 124.99,
|
| 1549 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1550 |
+
"eval_loss": 0.6857044696807861,
|
| 1551 |
+
"eval_runtime": 5.6134,
|
| 1552 |
+
"eval_samples_per_second": 50.95,
|
| 1553 |
+
"eval_steps_per_second": 3.207,
|
| 1554 |
+
"step": 3031
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 125.77,
|
| 1558 |
+
"grad_norm": 1.7039302587509155,
|
| 1559 |
+
"learning_rate": 1.6394444444444446e-05,
|
| 1560 |
+
"loss": 0.3101,
|
| 1561 |
+
"step": 3050
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 125.98,
|
| 1565 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1566 |
+
"eval_loss": 0.6742061972618103,
|
| 1567 |
+
"eval_runtime": 5.3609,
|
| 1568 |
+
"eval_samples_per_second": 53.349,
|
| 1569 |
+
"eval_steps_per_second": 3.358,
|
| 1570 |
+
"step": 3055
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 126.97,
|
| 1574 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1575 |
+
"eval_loss": 0.7029407620429993,
|
| 1576 |
+
"eval_runtime": 5.8891,
|
| 1577 |
+
"eval_samples_per_second": 48.564,
|
| 1578 |
+
"eval_steps_per_second": 3.056,
|
| 1579 |
+
"step": 3079
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 127.84,
|
| 1583 |
+
"grad_norm": 1.434970736503601,
|
| 1584 |
+
"learning_rate": 1.6116666666666668e-05,
|
| 1585 |
+
"loss": 0.284,
|
| 1586 |
+
"step": 3100
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 128.0,
|
| 1590 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1591 |
+
"eval_loss": 0.682050347328186,
|
| 1592 |
+
"eval_runtime": 5.1437,
|
| 1593 |
+
"eval_samples_per_second": 55.602,
|
| 1594 |
+
"eval_steps_per_second": 3.499,
|
| 1595 |
+
"step": 3104
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 128.99,
|
| 1599 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1600 |
+
"eval_loss": 0.68370121717453,
|
| 1601 |
+
"eval_runtime": 4.2733,
|
| 1602 |
+
"eval_samples_per_second": 66.927,
|
| 1603 |
+
"eval_steps_per_second": 4.212,
|
| 1604 |
+
"step": 3128
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 129.9,
|
| 1608 |
+
"grad_norm": 1.320789098739624,
|
| 1609 |
+
"learning_rate": 1.583888888888889e-05,
|
| 1610 |
+
"loss": 0.2902,
|
| 1611 |
+
"step": 3150
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 129.98,
|
| 1615 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1616 |
+
"eval_loss": 0.6823462843894958,
|
| 1617 |
+
"eval_runtime": 5.7566,
|
| 1618 |
+
"eval_samples_per_second": 49.682,
|
| 1619 |
+
"eval_steps_per_second": 3.127,
|
| 1620 |
+
"step": 3152
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 130.97,
|
| 1624 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1625 |
+
"eval_loss": 0.6950440406799316,
|
| 1626 |
+
"eval_runtime": 4.9248,
|
| 1627 |
+
"eval_samples_per_second": 58.074,
|
| 1628 |
+
"eval_steps_per_second": 3.655,
|
| 1629 |
+
"step": 3176
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 131.96,
|
| 1633 |
+
"grad_norm": 2.1280930042266846,
|
| 1634 |
+
"learning_rate": 1.556111111111111e-05,
|
| 1635 |
+
"loss": 0.301,
|
| 1636 |
+
"step": 3200
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 132.0,
|
| 1640 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1641 |
+
"eval_loss": 0.6800761818885803,
|
| 1642 |
+
"eval_runtime": 8.1328,
|
| 1643 |
+
"eval_samples_per_second": 35.166,
|
| 1644 |
+
"eval_steps_per_second": 2.213,
|
| 1645 |
+
"step": 3201
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 132.99,
|
| 1649 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1650 |
+
"eval_loss": 0.6867505311965942,
|
| 1651 |
+
"eval_runtime": 4.2532,
|
| 1652 |
+
"eval_samples_per_second": 67.244,
|
| 1653 |
+
"eval_steps_per_second": 4.232,
|
| 1654 |
+
"step": 3225
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 133.98,
|
| 1658 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1659 |
+
"eval_loss": 0.7061284184455872,
|
| 1660 |
+
"eval_runtime": 5.3031,
|
| 1661 |
+
"eval_samples_per_second": 53.93,
|
| 1662 |
+
"eval_steps_per_second": 3.394,
|
| 1663 |
+
"step": 3249
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 134.02,
|
| 1667 |
+
"grad_norm": 1.532638669013977,
|
| 1668 |
+
"learning_rate": 1.5283333333333333e-05,
|
| 1669 |
+
"loss": 0.2736,
|
| 1670 |
+
"step": 3250
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 134.97,
|
| 1674 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1675 |
+
"eval_loss": 0.7114368677139282,
|
| 1676 |
+
"eval_runtime": 4.6536,
|
| 1677 |
+
"eval_samples_per_second": 61.458,
|
| 1678 |
+
"eval_steps_per_second": 3.868,
|
| 1679 |
+
"step": 3273
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 136.0,
|
| 1683 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1684 |
+
"eval_loss": 0.6914551854133606,
|
| 1685 |
+
"eval_runtime": 4.5505,
|
| 1686 |
+
"eval_samples_per_second": 62.851,
|
| 1687 |
+
"eval_steps_per_second": 3.956,
|
| 1688 |
+
"step": 3298
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 136.08,
|
| 1692 |
+
"grad_norm": 2.0108492374420166,
|
| 1693 |
+
"learning_rate": 1.5005555555555555e-05,
|
| 1694 |
+
"loss": 0.2931,
|
| 1695 |
+
"step": 3300
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 136.99,
|
| 1699 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1700 |
+
"eval_loss": 0.7055917978286743,
|
| 1701 |
+
"eval_runtime": 5.3067,
|
| 1702 |
+
"eval_samples_per_second": 53.894,
|
| 1703 |
+
"eval_steps_per_second": 3.392,
|
| 1704 |
+
"step": 3322
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 137.98,
|
| 1708 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1709 |
+
"eval_loss": 0.7026935815811157,
|
| 1710 |
+
"eval_runtime": 5.186,
|
| 1711 |
+
"eval_samples_per_second": 55.149,
|
| 1712 |
+
"eval_steps_per_second": 3.471,
|
| 1713 |
+
"step": 3346
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 138.14,
|
| 1717 |
+
"grad_norm": 1.0804469585418701,
|
| 1718 |
+
"learning_rate": 1.4727777777777779e-05,
|
| 1719 |
+
"loss": 0.2864,
|
| 1720 |
+
"step": 3350
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 138.97,
|
| 1724 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1725 |
+
"eval_loss": 0.6983500719070435,
|
| 1726 |
+
"eval_runtime": 6.955,
|
| 1727 |
+
"eval_samples_per_second": 41.122,
|
| 1728 |
+
"eval_steps_per_second": 2.588,
|
| 1729 |
+
"step": 3370
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 140.0,
|
| 1733 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1734 |
+
"eval_loss": 0.7168787121772766,
|
| 1735 |
+
"eval_runtime": 4.234,
|
| 1736 |
+
"eval_samples_per_second": 67.548,
|
| 1737 |
+
"eval_steps_per_second": 4.251,
|
| 1738 |
+
"step": 3395
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 140.21,
|
| 1742 |
+
"grad_norm": 2.370694637298584,
|
| 1743 |
+
"learning_rate": 1.445e-05,
|
| 1744 |
+
"loss": 0.2765,
|
| 1745 |
+
"step": 3400
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 140.99,
|
| 1749 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1750 |
+
"eval_loss": 0.6960318088531494,
|
| 1751 |
+
"eval_runtime": 5.0294,
|
| 1752 |
+
"eval_samples_per_second": 56.865,
|
| 1753 |
+
"eval_steps_per_second": 3.579,
|
| 1754 |
+
"step": 3419
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 141.98,
|
| 1758 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1759 |
+
"eval_loss": 0.6990492343902588,
|
| 1760 |
+
"eval_runtime": 5.2727,
|
| 1761 |
+
"eval_samples_per_second": 54.242,
|
| 1762 |
+
"eval_steps_per_second": 3.414,
|
| 1763 |
+
"step": 3443
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 142.27,
|
| 1767 |
+
"grad_norm": 1.6676194667816162,
|
| 1768 |
+
"learning_rate": 1.4172222222222222e-05,
|
| 1769 |
+
"loss": 0.2808,
|
| 1770 |
+
"step": 3450
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 142.97,
|
| 1774 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1775 |
+
"eval_loss": 0.706200897693634,
|
| 1776 |
+
"eval_runtime": 4.5273,
|
| 1777 |
+
"eval_samples_per_second": 63.173,
|
| 1778 |
+
"eval_steps_per_second": 3.976,
|
| 1779 |
+
"step": 3467
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 144.0,
|
| 1783 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1784 |
+
"eval_loss": 0.6821764707565308,
|
| 1785 |
+
"eval_runtime": 5.3614,
|
| 1786 |
+
"eval_samples_per_second": 53.344,
|
| 1787 |
+
"eval_steps_per_second": 3.357,
|
| 1788 |
+
"step": 3492
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 144.33,
|
| 1792 |
+
"grad_norm": 1.9151145219802856,
|
| 1793 |
+
"learning_rate": 1.3894444444444444e-05,
|
| 1794 |
+
"loss": 0.2712,
|
| 1795 |
+
"step": 3500
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 144.99,
|
| 1799 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1800 |
+
"eval_loss": 0.7063603401184082,
|
| 1801 |
+
"eval_runtime": 4.9088,
|
| 1802 |
+
"eval_samples_per_second": 58.263,
|
| 1803 |
+
"eval_steps_per_second": 3.667,
|
| 1804 |
+
"step": 3516
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 145.98,
|
| 1808 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1809 |
+
"eval_loss": 0.7150112390518188,
|
| 1810 |
+
"eval_runtime": 7.2044,
|
| 1811 |
+
"eval_samples_per_second": 39.698,
|
| 1812 |
+
"eval_steps_per_second": 2.498,
|
| 1813 |
+
"step": 3540
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 146.39,
|
| 1817 |
+
"grad_norm": 1.5093848705291748,
|
| 1818 |
+
"learning_rate": 1.3622222222222223e-05,
|
| 1819 |
+
"loss": 0.2726,
|
| 1820 |
+
"step": 3550
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 146.97,
|
| 1824 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1825 |
+
"eval_loss": 0.696849524974823,
|
| 1826 |
+
"eval_runtime": 4.9386,
|
| 1827 |
+
"eval_samples_per_second": 57.911,
|
| 1828 |
+
"eval_steps_per_second": 3.645,
|
| 1829 |
+
"step": 3564
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 148.0,
|
| 1833 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1834 |
+
"eval_loss": 0.7086759209632874,
|
| 1835 |
+
"eval_runtime": 4.4363,
|
| 1836 |
+
"eval_samples_per_second": 64.468,
|
| 1837 |
+
"eval_steps_per_second": 4.057,
|
| 1838 |
+
"step": 3589
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 148.45,
|
| 1842 |
+
"grad_norm": 1.4403679370880127,
|
| 1843 |
+
"learning_rate": 1.3344444444444444e-05,
|
| 1844 |
+
"loss": 0.2607,
|
| 1845 |
+
"step": 3600
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 148.99,
|
| 1849 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1850 |
+
"eval_loss": 0.7129560112953186,
|
| 1851 |
+
"eval_runtime": 5.3809,
|
| 1852 |
+
"eval_samples_per_second": 53.15,
|
| 1853 |
+
"eval_steps_per_second": 3.345,
|
| 1854 |
+
"step": 3613
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 149.98,
|
| 1858 |
+
"eval_accuracy": 0.7902097902097902,
|
| 1859 |
+
"eval_loss": 0.7080287933349609,
|
| 1860 |
+
"eval_runtime": 5.8187,
|
| 1861 |
+
"eval_samples_per_second": 49.152,
|
| 1862 |
+
"eval_steps_per_second": 3.093,
|
| 1863 |
+
"step": 3637
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 150.52,
|
| 1867 |
+
"grad_norm": 2.036515235900879,
|
| 1868 |
+
"learning_rate": 1.3066666666666666e-05,
|
| 1869 |
+
"loss": 0.2546,
|
| 1870 |
+
"step": 3650
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 150.97,
|
| 1874 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1875 |
+
"eval_loss": 0.7088435888290405,
|
| 1876 |
+
"eval_runtime": 4.8742,
|
| 1877 |
+
"eval_samples_per_second": 58.677,
|
| 1878 |
+
"eval_steps_per_second": 3.693,
|
| 1879 |
+
"step": 3661
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 152.0,
|
| 1883 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1884 |
+
"eval_loss": 0.7030193209648132,
|
| 1885 |
+
"eval_runtime": 4.9492,
|
| 1886 |
+
"eval_samples_per_second": 57.787,
|
| 1887 |
+
"eval_steps_per_second": 3.637,
|
| 1888 |
+
"step": 3686
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 152.58,
|
| 1892 |
+
"grad_norm": 1.200052261352539,
|
| 1893 |
+
"learning_rate": 1.2788888888888888e-05,
|
| 1894 |
+
"loss": 0.2563,
|
| 1895 |
+
"step": 3700
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 152.99,
|
| 1899 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1900 |
+
"eval_loss": 0.7077969908714294,
|
| 1901 |
+
"eval_runtime": 4.614,
|
| 1902 |
+
"eval_samples_per_second": 61.985,
|
| 1903 |
+
"eval_steps_per_second": 3.901,
|
| 1904 |
+
"step": 3710
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 153.98,
|
| 1908 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1909 |
+
"eval_loss": 0.700455904006958,
|
| 1910 |
+
"eval_runtime": 5.7657,
|
| 1911 |
+
"eval_samples_per_second": 49.604,
|
| 1912 |
+
"eval_steps_per_second": 3.122,
|
| 1913 |
+
"step": 3734
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 154.64,
|
| 1917 |
+
"grad_norm": 2.2751214504241943,
|
| 1918 |
+
"learning_rate": 1.2511111111111112e-05,
|
| 1919 |
+
"loss": 0.2531,
|
| 1920 |
+
"step": 3750
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 154.97,
|
| 1924 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1925 |
+
"eval_loss": 0.7160292267799377,
|
| 1926 |
+
"eval_runtime": 5.1079,
|
| 1927 |
+
"eval_samples_per_second": 55.992,
|
| 1928 |
+
"eval_steps_per_second": 3.524,
|
| 1929 |
+
"step": 3758
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 156.0,
|
| 1933 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1934 |
+
"eval_loss": 0.7175909876823425,
|
| 1935 |
+
"eval_runtime": 5.4035,
|
| 1936 |
+
"eval_samples_per_second": 52.929,
|
| 1937 |
+
"eval_steps_per_second": 3.331,
|
| 1938 |
+
"step": 3783
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 156.7,
|
| 1942 |
+
"grad_norm": 1.9024412631988525,
|
| 1943 |
+
"learning_rate": 1.2233333333333334e-05,
|
| 1944 |
+
"loss": 0.2446,
|
| 1945 |
+
"step": 3800
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 156.99,
|
| 1949 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1950 |
+
"eval_loss": 0.7190600037574768,
|
| 1951 |
+
"eval_runtime": 4.3633,
|
| 1952 |
+
"eval_samples_per_second": 65.546,
|
| 1953 |
+
"eval_steps_per_second": 4.125,
|
| 1954 |
+
"step": 3807
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 157.98,
|
| 1958 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1959 |
+
"eval_loss": 0.719641387462616,
|
| 1960 |
+
"eval_runtime": 5.0426,
|
| 1961 |
+
"eval_samples_per_second": 56.717,
|
| 1962 |
+
"eval_steps_per_second": 3.57,
|
| 1963 |
+
"step": 3831
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 158.76,
|
| 1967 |
+
"grad_norm": 3.471806287765503,
|
| 1968 |
+
"learning_rate": 1.1955555555555556e-05,
|
| 1969 |
+
"loss": 0.2479,
|
| 1970 |
+
"step": 3850
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 158.97,
|
| 1974 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1975 |
+
"eval_loss": 0.7073430418968201,
|
| 1976 |
+
"eval_runtime": 3.6336,
|
| 1977 |
+
"eval_samples_per_second": 78.711,
|
| 1978 |
+
"eval_steps_per_second": 4.954,
|
| 1979 |
+
"step": 3855
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 160.0,
|
| 1983 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1984 |
+
"eval_loss": 0.7328661680221558,
|
| 1985 |
+
"eval_runtime": 5.2625,
|
| 1986 |
+
"eval_samples_per_second": 54.347,
|
| 1987 |
+
"eval_steps_per_second": 3.42,
|
| 1988 |
+
"step": 3880
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 160.82,
|
| 1992 |
+
"grad_norm": 2.1171793937683105,
|
| 1993 |
+
"learning_rate": 1.1677777777777777e-05,
|
| 1994 |
+
"loss": 0.2523,
|
| 1995 |
+
"step": 3900
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 160.99,
|
| 1999 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2000 |
+
"eval_loss": 0.7158821821212769,
|
| 2001 |
+
"eval_runtime": 6.5877,
|
| 2002 |
+
"eval_samples_per_second": 43.414,
|
| 2003 |
+
"eval_steps_per_second": 2.732,
|
| 2004 |
+
"step": 3904
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 161.98,
|
| 2008 |
+
"eval_accuracy": 0.7692307692307693,
|
| 2009 |
+
"eval_loss": 0.719171404838562,
|
| 2010 |
+
"eval_runtime": 4.5674,
|
| 2011 |
+
"eval_samples_per_second": 62.618,
|
| 2012 |
+
"eval_steps_per_second": 3.941,
|
| 2013 |
+
"step": 3928
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 162.89,
|
| 2017 |
+
"grad_norm": 1.7515395879745483,
|
| 2018 |
+
"learning_rate": 1.1400000000000001e-05,
|
| 2019 |
+
"loss": 0.2523,
|
| 2020 |
+
"step": 3950
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 162.97,
|
| 2024 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2025 |
+
"eval_loss": 0.7281435132026672,
|
| 2026 |
+
"eval_runtime": 4.4866,
|
| 2027 |
+
"eval_samples_per_second": 63.746,
|
| 2028 |
+
"eval_steps_per_second": 4.012,
|
| 2029 |
+
"step": 3952
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 164.0,
|
| 2033 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2034 |
+
"eval_loss": 0.7078841328620911,
|
| 2035 |
+
"eval_runtime": 4.4241,
|
| 2036 |
+
"eval_samples_per_second": 64.645,
|
| 2037 |
+
"eval_steps_per_second": 4.069,
|
| 2038 |
+
"step": 3977
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 164.95,
|
| 2042 |
+
"grad_norm": 1.456335186958313,
|
| 2043 |
+
"learning_rate": 1.1122222222222223e-05,
|
| 2044 |
+
"loss": 0.2422,
|
| 2045 |
+
"step": 4000
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 164.99,
|
| 2049 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2050 |
+
"eval_loss": 0.7161521911621094,
|
| 2051 |
+
"eval_runtime": 5.1239,
|
| 2052 |
+
"eval_samples_per_second": 55.817,
|
| 2053 |
+
"eval_steps_per_second": 3.513,
|
| 2054 |
+
"step": 4001
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 165.98,
|
| 2058 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2059 |
+
"eval_loss": 0.7190020084381104,
|
| 2060 |
+
"eval_runtime": 3.4488,
|
| 2061 |
+
"eval_samples_per_second": 82.926,
|
| 2062 |
+
"eval_steps_per_second": 5.219,
|
| 2063 |
+
"step": 4025
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 166.97,
|
| 2067 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2068 |
+
"eval_loss": 0.7311248779296875,
|
| 2069 |
+
"eval_runtime": 5.0389,
|
| 2070 |
+
"eval_samples_per_second": 56.759,
|
| 2071 |
+
"eval_steps_per_second": 3.572,
|
| 2072 |
+
"step": 4049
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 167.01,
|
| 2076 |
+
"grad_norm": 1.2554075717926025,
|
| 2077 |
+
"learning_rate": 1.0844444444444445e-05,
|
| 2078 |
+
"loss": 0.242,
|
| 2079 |
+
"step": 4050
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 168.0,
|
| 2083 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2084 |
+
"eval_loss": 0.7110462188720703,
|
| 2085 |
+
"eval_runtime": 4.4612,
|
| 2086 |
+
"eval_samples_per_second": 64.108,
|
| 2087 |
+
"eval_steps_per_second": 4.035,
|
| 2088 |
+
"step": 4074
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 168.99,
|
| 2092 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2093 |
+
"eval_loss": 0.7028501629829407,
|
| 2094 |
+
"eval_runtime": 6.955,
|
| 2095 |
+
"eval_samples_per_second": 41.122,
|
| 2096 |
+
"eval_steps_per_second": 2.588,
|
| 2097 |
+
"step": 4098
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 169.07,
|
| 2101 |
+
"grad_norm": 2.8003265857696533,
|
| 2102 |
+
"learning_rate": 1.0566666666666667e-05,
|
| 2103 |
+
"loss": 0.2392,
|
| 2104 |
+
"step": 4100
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 169.98,
|
| 2108 |
+
"eval_accuracy": 0.7937062937062938,
|
| 2109 |
+
"eval_loss": 0.7108554840087891,
|
| 2110 |
+
"eval_runtime": 5.0033,
|
| 2111 |
+
"eval_samples_per_second": 57.162,
|
| 2112 |
+
"eval_steps_per_second": 3.598,
|
| 2113 |
+
"step": 4122
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 170.97,
|
| 2117 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2118 |
+
"eval_loss": 0.7106384634971619,
|
| 2119 |
+
"eval_runtime": 5.1984,
|
| 2120 |
+
"eval_samples_per_second": 55.017,
|
| 2121 |
+
"eval_steps_per_second": 3.463,
|
| 2122 |
+
"step": 4146
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 171.13,
|
| 2126 |
+
"grad_norm": 2.1897969245910645,
|
| 2127 |
+
"learning_rate": 1.028888888888889e-05,
|
| 2128 |
+
"loss": 0.247,
|
| 2129 |
+
"step": 4150
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 172.0,
|
| 2133 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2134 |
+
"eval_loss": 0.7151694297790527,
|
| 2135 |
+
"eval_runtime": 5.1963,
|
| 2136 |
+
"eval_samples_per_second": 55.039,
|
| 2137 |
+
"eval_steps_per_second": 3.464,
|
| 2138 |
+
"step": 4171
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 172.99,
|
| 2142 |
+
"eval_accuracy": 0.7657342657342657,
|
| 2143 |
+
"eval_loss": 0.7254167795181274,
|
| 2144 |
+
"eval_runtime": 4.4466,
|
| 2145 |
+
"eval_samples_per_second": 64.319,
|
| 2146 |
+
"eval_steps_per_second": 4.048,
|
| 2147 |
+
"step": 4195
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 173.2,
|
| 2151 |
+
"grad_norm": 2.769357681274414,
|
| 2152 |
+
"learning_rate": 1.0011111111111112e-05,
|
| 2153 |
+
"loss": 0.2341,
|
| 2154 |
+
"step": 4200
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 173.98,
|
| 2158 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2159 |
+
"eval_loss": 0.7290962338447571,
|
| 2160 |
+
"eval_runtime": 6.2221,
|
| 2161 |
+
"eval_samples_per_second": 45.965,
|
| 2162 |
+
"eval_steps_per_second": 2.893,
|
| 2163 |
+
"step": 4219
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 174.97,
|
| 2167 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2168 |
+
"eval_loss": 0.7088623046875,
|
| 2169 |
+
"eval_runtime": 4.3709,
|
| 2170 |
+
"eval_samples_per_second": 65.433,
|
| 2171 |
+
"eval_steps_per_second": 4.118,
|
| 2172 |
+
"step": 4243
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 175.26,
|
| 2176 |
+
"grad_norm": 2.044703483581543,
|
| 2177 |
+
"learning_rate": 9.733333333333332e-06,
|
| 2178 |
+
"loss": 0.2317,
|
| 2179 |
+
"step": 4250
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 176.0,
|
| 2183 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2184 |
+
"eval_loss": 0.7185826897621155,
|
| 2185 |
+
"eval_runtime": 5.4095,
|
| 2186 |
+
"eval_samples_per_second": 52.87,
|
| 2187 |
+
"eval_steps_per_second": 3.327,
|
| 2188 |
+
"step": 4268
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 176.99,
|
| 2192 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2193 |
+
"eval_loss": 0.7167823314666748,
|
| 2194 |
+
"eval_runtime": 4.9506,
|
| 2195 |
+
"eval_samples_per_second": 57.77,
|
| 2196 |
+
"eval_steps_per_second": 3.636,
|
| 2197 |
+
"step": 4292
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"epoch": 177.32,
|
| 2201 |
+
"grad_norm": 1.078834056854248,
|
| 2202 |
+
"learning_rate": 9.455555555555556e-06,
|
| 2203 |
+
"loss": 0.2269,
|
| 2204 |
+
"step": 4300
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 177.98,
|
| 2208 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2209 |
+
"eval_loss": 0.7237738966941833,
|
| 2210 |
+
"eval_runtime": 4.781,
|
| 2211 |
+
"eval_samples_per_second": 59.82,
|
| 2212 |
+
"eval_steps_per_second": 3.765,
|
| 2213 |
+
"step": 4316
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 178.97,
|
| 2217 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2218 |
+
"eval_loss": 0.7131801247596741,
|
| 2219 |
+
"eval_runtime": 4.6869,
|
| 2220 |
+
"eval_samples_per_second": 61.022,
|
| 2221 |
+
"eval_steps_per_second": 3.841,
|
| 2222 |
+
"step": 4340
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 179.38,
|
| 2226 |
+
"grad_norm": 2.008120536804199,
|
| 2227 |
+
"learning_rate": 9.177777777777778e-06,
|
| 2228 |
+
"loss": 0.2283,
|
| 2229 |
+
"step": 4350
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 180.0,
|
| 2233 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2234 |
+
"eval_loss": 0.7384253144264221,
|
| 2235 |
+
"eval_runtime": 4.5879,
|
| 2236 |
+
"eval_samples_per_second": 62.338,
|
| 2237 |
+
"eval_steps_per_second": 3.923,
|
| 2238 |
+
"step": 4365
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 180.99,
|
| 2242 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2243 |
+
"eval_loss": 0.7002861499786377,
|
| 2244 |
+
"eval_runtime": 5.3238,
|
| 2245 |
+
"eval_samples_per_second": 53.721,
|
| 2246 |
+
"eval_steps_per_second": 3.381,
|
| 2247 |
+
"step": 4389
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"epoch": 181.44,
|
| 2251 |
+
"grad_norm": 1.9518792629241943,
|
| 2252 |
+
"learning_rate": 8.900000000000001e-06,
|
| 2253 |
+
"loss": 0.2303,
|
| 2254 |
+
"step": 4400
|
| 2255 |
+
},
|
| 2256 |
+
{
|
| 2257 |
+
"epoch": 181.98,
|
| 2258 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2259 |
+
"eval_loss": 0.7278482913970947,
|
| 2260 |
+
"eval_runtime": 5.8358,
|
| 2261 |
+
"eval_samples_per_second": 49.008,
|
| 2262 |
+
"eval_steps_per_second": 3.084,
|
| 2263 |
+
"step": 4413
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 182.97,
|
| 2267 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2268 |
+
"eval_loss": 0.7143127918243408,
|
| 2269 |
+
"eval_runtime": 6.1229,
|
| 2270 |
+
"eval_samples_per_second": 46.71,
|
| 2271 |
+
"eval_steps_per_second": 2.94,
|
| 2272 |
+
"step": 4437
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 183.51,
|
| 2276 |
+
"grad_norm": 1.0936890840530396,
|
| 2277 |
+
"learning_rate": 8.622222222222221e-06,
|
| 2278 |
+
"loss": 0.2109,
|
| 2279 |
+
"step": 4450
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 184.0,
|
| 2283 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2284 |
+
"eval_loss": 0.7406834363937378,
|
| 2285 |
+
"eval_runtime": 5.0467,
|
| 2286 |
+
"eval_samples_per_second": 56.671,
|
| 2287 |
+
"eval_steps_per_second": 3.567,
|
| 2288 |
+
"step": 4462
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 184.99,
|
| 2292 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2293 |
+
"eval_loss": 0.7053534388542175,
|
| 2294 |
+
"eval_runtime": 5.279,
|
| 2295 |
+
"eval_samples_per_second": 54.177,
|
| 2296 |
+
"eval_steps_per_second": 3.41,
|
| 2297 |
+
"step": 4486
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 185.57,
|
| 2301 |
+
"grad_norm": 2.9350059032440186,
|
| 2302 |
+
"learning_rate": 8.344444444444445e-06,
|
| 2303 |
+
"loss": 0.2261,
|
| 2304 |
+
"step": 4500
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 185.98,
|
| 2308 |
+
"eval_accuracy": 0.7727272727272727,
|
| 2309 |
+
"eval_loss": 0.7260809540748596,
|
| 2310 |
+
"eval_runtime": 5.4165,
|
| 2311 |
+
"eval_samples_per_second": 52.802,
|
| 2312 |
+
"eval_steps_per_second": 3.323,
|
| 2313 |
+
"step": 4510
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 186.97,
|
| 2317 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2318 |
+
"eval_loss": 0.7240064144134521,
|
| 2319 |
+
"eval_runtime": 5.4866,
|
| 2320 |
+
"eval_samples_per_second": 52.127,
|
| 2321 |
+
"eval_steps_per_second": 3.281,
|
| 2322 |
+
"step": 4534
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 187.63,
|
| 2326 |
+
"grad_norm": 1.8322782516479492,
|
| 2327 |
+
"learning_rate": 8.066666666666667e-06,
|
| 2328 |
+
"loss": 0.2282,
|
| 2329 |
+
"step": 4550
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 188.0,
|
| 2333 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2334 |
+
"eval_loss": 0.7199599146842957,
|
| 2335 |
+
"eval_runtime": 4.6736,
|
| 2336 |
+
"eval_samples_per_second": 61.195,
|
| 2337 |
+
"eval_steps_per_second": 3.851,
|
| 2338 |
+
"step": 4559
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 188.99,
|
| 2342 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2343 |
+
"eval_loss": 0.7102844715118408,
|
| 2344 |
+
"eval_runtime": 5.4219,
|
| 2345 |
+
"eval_samples_per_second": 52.749,
|
| 2346 |
+
"eval_steps_per_second": 3.32,
|
| 2347 |
+
"step": 4583
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 189.69,
|
| 2351 |
+
"grad_norm": 1.8777916431427002,
|
| 2352 |
+
"learning_rate": 7.78888888888889e-06,
|
| 2353 |
+
"loss": 0.2321,
|
| 2354 |
+
"step": 4600
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 189.98,
|
| 2358 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2359 |
+
"eval_loss": 0.7083376049995422,
|
| 2360 |
+
"eval_runtime": 5.9634,
|
| 2361 |
+
"eval_samples_per_second": 47.959,
|
| 2362 |
+
"eval_steps_per_second": 3.018,
|
| 2363 |
+
"step": 4607
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 190.97,
|
| 2367 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2368 |
+
"eval_loss": 0.7244677543640137,
|
| 2369 |
+
"eval_runtime": 5.2078,
|
| 2370 |
+
"eval_samples_per_second": 54.918,
|
| 2371 |
+
"eval_steps_per_second": 3.456,
|
| 2372 |
+
"step": 4631
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 191.75,
|
| 2376 |
+
"grad_norm": 1.5277408361434937,
|
| 2377 |
+
"learning_rate": 7.5111111111111105e-06,
|
| 2378 |
+
"loss": 0.2261,
|
| 2379 |
+
"step": 4650
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 192.0,
|
| 2383 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2384 |
+
"eval_loss": 0.7124583721160889,
|
| 2385 |
+
"eval_runtime": 5.7079,
|
| 2386 |
+
"eval_samples_per_second": 50.106,
|
| 2387 |
+
"eval_steps_per_second": 3.154,
|
| 2388 |
+
"step": 4656
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 192.99,
|
| 2392 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2393 |
+
"eval_loss": 0.7308976054191589,
|
| 2394 |
+
"eval_runtime": 5.3404,
|
| 2395 |
+
"eval_samples_per_second": 53.554,
|
| 2396 |
+
"eval_steps_per_second": 3.371,
|
| 2397 |
+
"step": 4680
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 193.81,
|
| 2401 |
+
"grad_norm": 2.095749616622925,
|
| 2402 |
+
"learning_rate": 7.233333333333333e-06,
|
| 2403 |
+
"loss": 0.2231,
|
| 2404 |
+
"step": 4700
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 193.98,
|
| 2408 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2409 |
+
"eval_loss": 0.7237818837165833,
|
| 2410 |
+
"eval_runtime": 4.6666,
|
| 2411 |
+
"eval_samples_per_second": 61.286,
|
| 2412 |
+
"eval_steps_per_second": 3.857,
|
| 2413 |
+
"step": 4704
|
| 2414 |
+
},
|
| 2415 |
+
{
|
| 2416 |
+
"epoch": 194.97,
|
| 2417 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2418 |
+
"eval_loss": 0.7253320217132568,
|
| 2419 |
+
"eval_runtime": 5.8059,
|
| 2420 |
+
"eval_samples_per_second": 49.261,
|
| 2421 |
+
"eval_steps_per_second": 3.1,
|
| 2422 |
+
"step": 4728
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 195.88,
|
| 2426 |
+
"grad_norm": 1.6955636739730835,
|
| 2427 |
+
"learning_rate": 6.955555555555556e-06,
|
| 2428 |
+
"loss": 0.2083,
|
| 2429 |
+
"step": 4750
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 196.0,
|
| 2433 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2434 |
+
"eval_loss": 0.7240011692047119,
|
| 2435 |
+
"eval_runtime": 6.0767,
|
| 2436 |
+
"eval_samples_per_second": 47.065,
|
| 2437 |
+
"eval_steps_per_second": 2.962,
|
| 2438 |
+
"step": 4753
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 196.99,
|
| 2442 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2443 |
+
"eval_loss": 0.7131750583648682,
|
| 2444 |
+
"eval_runtime": 5.3063,
|
| 2445 |
+
"eval_samples_per_second": 53.898,
|
| 2446 |
+
"eval_steps_per_second": 3.392,
|
| 2447 |
+
"step": 4777
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 197.94,
|
| 2451 |
+
"grad_norm": 0.8933289051055908,
|
| 2452 |
+
"learning_rate": 6.677777777777778e-06,
|
| 2453 |
+
"loss": 0.2116,
|
| 2454 |
+
"step": 4800
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 197.98,
|
| 2458 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2459 |
+
"eval_loss": 0.7169559597969055,
|
| 2460 |
+
"eval_runtime": 5.5713,
|
| 2461 |
+
"eval_samples_per_second": 51.335,
|
| 2462 |
+
"eval_steps_per_second": 3.231,
|
| 2463 |
+
"step": 4801
|
| 2464 |
+
},
|
| 2465 |
+
{
|
| 2466 |
+
"epoch": 198.97,
|
| 2467 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2468 |
+
"eval_loss": 0.7265609502792358,
|
| 2469 |
+
"eval_runtime": 4.1397,
|
| 2470 |
+
"eval_samples_per_second": 69.087,
|
| 2471 |
+
"eval_steps_per_second": 4.348,
|
| 2472 |
+
"step": 4825
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 200.0,
|
| 2476 |
+
"grad_norm": 2.175414562225342,
|
| 2477 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 2478 |
+
"loss": 0.2219,
|
| 2479 |
+
"step": 4850
|
| 2480 |
+
},
|
| 2481 |
+
{
|
| 2482 |
+
"epoch": 200.0,
|
| 2483 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2484 |
+
"eval_loss": 0.7162622213363647,
|
| 2485 |
+
"eval_runtime": 5.2016,
|
| 2486 |
+
"eval_samples_per_second": 54.984,
|
| 2487 |
+
"eval_steps_per_second": 3.461,
|
| 2488 |
+
"step": 4850
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 200.99,
|
| 2492 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2493 |
+
"eval_loss": 0.7302048802375793,
|
| 2494 |
+
"eval_runtime": 4.9222,
|
| 2495 |
+
"eval_samples_per_second": 58.104,
|
| 2496 |
+
"eval_steps_per_second": 3.657,
|
| 2497 |
+
"step": 4874
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 201.98,
|
| 2501 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2502 |
+
"eval_loss": 0.7223746180534363,
|
| 2503 |
+
"eval_runtime": 4.6884,
|
| 2504 |
+
"eval_samples_per_second": 61.002,
|
| 2505 |
+
"eval_steps_per_second": 3.839,
|
| 2506 |
+
"step": 4898
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 202.06,
|
| 2510 |
+
"grad_norm": 2.053739309310913,
|
| 2511 |
+
"learning_rate": 6.1222222222222224e-06,
|
| 2512 |
+
"loss": 0.2183,
|
| 2513 |
+
"step": 4900
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 202.97,
|
| 2517 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2518 |
+
"eval_loss": 0.7179226279258728,
|
| 2519 |
+
"eval_runtime": 4.5556,
|
| 2520 |
+
"eval_samples_per_second": 62.78,
|
| 2521 |
+
"eval_steps_per_second": 3.951,
|
| 2522 |
+
"step": 4922
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 204.0,
|
| 2526 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2527 |
+
"eval_loss": 0.7245286107063293,
|
| 2528 |
+
"eval_runtime": 5.7474,
|
| 2529 |
+
"eval_samples_per_second": 49.762,
|
| 2530 |
+
"eval_steps_per_second": 3.132,
|
| 2531 |
+
"step": 4947
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 204.12,
|
| 2535 |
+
"grad_norm": 1.1081063747406006,
|
| 2536 |
+
"learning_rate": 5.844444444444444e-06,
|
| 2537 |
+
"loss": 0.2053,
|
| 2538 |
+
"step": 4950
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 204.99,
|
| 2542 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2543 |
+
"eval_loss": 0.7344977259635925,
|
| 2544 |
+
"eval_runtime": 5.4178,
|
| 2545 |
+
"eval_samples_per_second": 52.789,
|
| 2546 |
+
"eval_steps_per_second": 3.322,
|
| 2547 |
+
"step": 4971
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 205.98,
|
| 2551 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2552 |
+
"eval_loss": 0.7249557971954346,
|
| 2553 |
+
"eval_runtime": 5.6352,
|
| 2554 |
+
"eval_samples_per_second": 50.753,
|
| 2555 |
+
"eval_steps_per_second": 3.194,
|
| 2556 |
+
"step": 4995
|
| 2557 |
+
},
|
| 2558 |
+
{
|
| 2559 |
+
"epoch": 206.19,
|
| 2560 |
+
"grad_norm": 1.09213125705719,
|
| 2561 |
+
"learning_rate": 5.566666666666667e-06,
|
| 2562 |
+
"loss": 0.2113,
|
| 2563 |
+
"step": 5000
|
| 2564 |
+
},
|
| 2565 |
+
{
|
| 2566 |
+
"epoch": 206.97,
|
| 2567 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2568 |
+
"eval_loss": 0.7246001958847046,
|
| 2569 |
+
"eval_runtime": 4.9071,
|
| 2570 |
+
"eval_samples_per_second": 58.283,
|
| 2571 |
+
"eval_steps_per_second": 3.668,
|
| 2572 |
+
"step": 5019
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 208.0,
|
| 2576 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2577 |
+
"eval_loss": 0.7270117998123169,
|
| 2578 |
+
"eval_runtime": 5.8385,
|
| 2579 |
+
"eval_samples_per_second": 48.985,
|
| 2580 |
+
"eval_steps_per_second": 3.083,
|
| 2581 |
+
"step": 5044
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 208.25,
|
| 2585 |
+
"grad_norm": 1.6693130731582642,
|
| 2586 |
+
"learning_rate": 5.288888888888889e-06,
|
| 2587 |
+
"loss": 0.2152,
|
| 2588 |
+
"step": 5050
|
| 2589 |
+
},
|
| 2590 |
+
{
|
| 2591 |
+
"epoch": 208.99,
|
| 2592 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2593 |
+
"eval_loss": 0.7285901308059692,
|
| 2594 |
+
"eval_runtime": 5.489,
|
| 2595 |
+
"eval_samples_per_second": 52.104,
|
| 2596 |
+
"eval_steps_per_second": 3.279,
|
| 2597 |
+
"step": 5068
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 209.98,
|
| 2601 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2602 |
+
"eval_loss": 0.7332947254180908,
|
| 2603 |
+
"eval_runtime": 5.3017,
|
| 2604 |
+
"eval_samples_per_second": 53.945,
|
| 2605 |
+
"eval_steps_per_second": 3.395,
|
| 2606 |
+
"step": 5092
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 210.31,
|
| 2610 |
+
"grad_norm": 2.0511515140533447,
|
| 2611 |
+
"learning_rate": 5.011111111111112e-06,
|
| 2612 |
+
"loss": 0.2129,
|
| 2613 |
+
"step": 5100
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 210.97,
|
| 2617 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2618 |
+
"eval_loss": 0.7307863831520081,
|
| 2619 |
+
"eval_runtime": 5.2991,
|
| 2620 |
+
"eval_samples_per_second": 53.971,
|
| 2621 |
+
"eval_steps_per_second": 3.397,
|
| 2622 |
+
"step": 5116
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 212.0,
|
| 2626 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2627 |
+
"eval_loss": 0.7176437973976135,
|
| 2628 |
+
"eval_runtime": 4.9452,
|
| 2629 |
+
"eval_samples_per_second": 57.834,
|
| 2630 |
+
"eval_steps_per_second": 3.64,
|
| 2631 |
+
"step": 5141
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"epoch": 212.37,
|
| 2635 |
+
"grad_norm": 1.8491023778915405,
|
| 2636 |
+
"learning_rate": 4.7333333333333335e-06,
|
| 2637 |
+
"loss": 0.2173,
|
| 2638 |
+
"step": 5150
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 212.99,
|
| 2642 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2643 |
+
"eval_loss": 0.7334882020950317,
|
| 2644 |
+
"eval_runtime": 4.9602,
|
| 2645 |
+
"eval_samples_per_second": 57.659,
|
| 2646 |
+
"eval_steps_per_second": 3.629,
|
| 2647 |
+
"step": 5165
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"epoch": 213.98,
|
| 2651 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2652 |
+
"eval_loss": 0.7268483638763428,
|
| 2653 |
+
"eval_runtime": 5.885,
|
| 2654 |
+
"eval_samples_per_second": 48.598,
|
| 2655 |
+
"eval_steps_per_second": 3.059,
|
| 2656 |
+
"step": 5189
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 214.43,
|
| 2660 |
+
"grad_norm": 1.2067769765853882,
|
| 2661 |
+
"learning_rate": 4.455555555555556e-06,
|
| 2662 |
+
"loss": 0.2042,
|
| 2663 |
+
"step": 5200
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 214.97,
|
| 2667 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2668 |
+
"eval_loss": 0.7299237847328186,
|
| 2669 |
+
"eval_runtime": 5.7645,
|
| 2670 |
+
"eval_samples_per_second": 49.614,
|
| 2671 |
+
"eval_steps_per_second": 3.123,
|
| 2672 |
+
"step": 5213
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 216.0,
|
| 2676 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2677 |
+
"eval_loss": 0.7360625863075256,
|
| 2678 |
+
"eval_runtime": 4.7143,
|
| 2679 |
+
"eval_samples_per_second": 60.667,
|
| 2680 |
+
"eval_steps_per_second": 3.818,
|
| 2681 |
+
"step": 5238
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 216.49,
|
| 2685 |
+
"grad_norm": 1.3863427639007568,
|
| 2686 |
+
"learning_rate": 4.177777777777777e-06,
|
| 2687 |
+
"loss": 0.2112,
|
| 2688 |
+
"step": 5250
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 216.99,
|
| 2692 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2693 |
+
"eval_loss": 0.723866879940033,
|
| 2694 |
+
"eval_runtime": 5.3445,
|
| 2695 |
+
"eval_samples_per_second": 53.513,
|
| 2696 |
+
"eval_steps_per_second": 3.368,
|
| 2697 |
+
"step": 5262
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 217.98,
|
| 2701 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2702 |
+
"eval_loss": 0.7252445220947266,
|
| 2703 |
+
"eval_runtime": 4.6314,
|
| 2704 |
+
"eval_samples_per_second": 61.753,
|
| 2705 |
+
"eval_steps_per_second": 3.887,
|
| 2706 |
+
"step": 5286
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 218.56,
|
| 2710 |
+
"grad_norm": 1.1177924871444702,
|
| 2711 |
+
"learning_rate": 3.9e-06,
|
| 2712 |
+
"loss": 0.2007,
|
| 2713 |
+
"step": 5300
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 218.97,
|
| 2717 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2718 |
+
"eval_loss": 0.719983696937561,
|
| 2719 |
+
"eval_runtime": 4.865,
|
| 2720 |
+
"eval_samples_per_second": 58.787,
|
| 2721 |
+
"eval_steps_per_second": 3.7,
|
| 2722 |
+
"step": 5310
|
| 2723 |
+
},
|
| 2724 |
+
{
|
| 2725 |
+
"epoch": 220.0,
|
| 2726 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2727 |
+
"eval_loss": 0.7195786237716675,
|
| 2728 |
+
"eval_runtime": 5.5422,
|
| 2729 |
+
"eval_samples_per_second": 51.604,
|
| 2730 |
+
"eval_steps_per_second": 3.248,
|
| 2731 |
+
"step": 5335
|
| 2732 |
+
},
|
| 2733 |
+
{
|
| 2734 |
+
"epoch": 220.62,
|
| 2735 |
+
"grad_norm": 1.413304090499878,
|
| 2736 |
+
"learning_rate": 3.6222222222222226e-06,
|
| 2737 |
+
"loss": 0.2163,
|
| 2738 |
+
"step": 5350
|
| 2739 |
+
},
|
| 2740 |
+
{
|
| 2741 |
+
"epoch": 220.99,
|
| 2742 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2743 |
+
"eval_loss": 0.7309580445289612,
|
| 2744 |
+
"eval_runtime": 5.2512,
|
| 2745 |
+
"eval_samples_per_second": 54.463,
|
| 2746 |
+
"eval_steps_per_second": 3.428,
|
| 2747 |
+
"step": 5359
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 221.98,
|
| 2751 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2752 |
+
"eval_loss": 0.7313971519470215,
|
| 2753 |
+
"eval_runtime": 5.1151,
|
| 2754 |
+
"eval_samples_per_second": 55.913,
|
| 2755 |
+
"eval_steps_per_second": 3.519,
|
| 2756 |
+
"step": 5383
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"epoch": 222.68,
|
| 2760 |
+
"grad_norm": 3.0471901893615723,
|
| 2761 |
+
"learning_rate": 3.3444444444444445e-06,
|
| 2762 |
+
"loss": 0.2141,
|
| 2763 |
+
"step": 5400
|
| 2764 |
+
},
|
| 2765 |
+
{
|
| 2766 |
+
"epoch": 222.97,
|
| 2767 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2768 |
+
"eval_loss": 0.727938175201416,
|
| 2769 |
+
"eval_runtime": 4.6405,
|
| 2770 |
+
"eval_samples_per_second": 61.631,
|
| 2771 |
+
"eval_steps_per_second": 3.879,
|
| 2772 |
+
"step": 5407
|
| 2773 |
+
},
|
| 2774 |
+
{
|
| 2775 |
+
"epoch": 224.0,
|
| 2776 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2777 |
+
"eval_loss": 0.725923478603363,
|
| 2778 |
+
"eval_runtime": 5.0906,
|
| 2779 |
+
"eval_samples_per_second": 56.182,
|
| 2780 |
+
"eval_steps_per_second": 3.536,
|
| 2781 |
+
"step": 5432
|
| 2782 |
+
}
|
| 2783 |
+
],
|
| 2784 |
+
"logging_steps": 50,
|
| 2785 |
+
"max_steps": 6000,
|
| 2786 |
+
"num_input_tokens_seen": 0,
|
| 2787 |
+
"num_train_epochs": 250,
|
| 2788 |
+
"save_steps": 500,
|
| 2789 |
+
"total_flos": 3.037085846065152e+18,
|
| 2790 |
+
"train_batch_size": 16,
|
| 2791 |
+
"trial_name": null,
|
| 2792 |
+
"trial_params": null
|
| 2793 |
+
}
|
checkpoint-5432/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3218c866c7b253f5d3295edcd44b4197864747f68600a16bb0f3d6f506131fb
|
| 3 |
+
size 4984
|
config.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "anderloh/Hugginhface-master-wav2vec-pretreined-5-class-train-test",
|
| 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 |
+
"Wav2Vec2ForSequenceClassification"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 128,
|
| 15 |
+
"codevector_dim": 128,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
256,
|
| 20 |
+
256,
|
| 21 |
+
256,
|
| 22 |
+
256,
|
| 23 |
+
256,
|
| 24 |
+
256,
|
| 25 |
+
256
|
| 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": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"finetuning_task": "audio-classification",
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 59 |
+
"hidden_dropout_prob": 0.0,
|
| 60 |
+
"hidden_size": 384,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "Helicopter",
|
| 63 |
+
"1": "Jet",
|
| 64 |
+
"2": "Racecar",
|
| 65 |
+
"3": "Rail",
|
| 66 |
+
"4": "Truck"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 1536,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"Helicopter": "0",
|
| 72 |
+
"Jet": "1",
|
| 73 |
+
"Racecar": "2",
|
| 74 |
+
"Rail": "3",
|
| 75 |
+
"Truck": "4"
|
| 76 |
+
},
|
| 77 |
+
"layer_norm_eps": 1e-05,
|
| 78 |
+
"layerdrop": 0.0,
|
| 79 |
+
"mask_feature_length": 10,
|
| 80 |
+
"mask_feature_min_masks": 0,
|
| 81 |
+
"mask_feature_prob": 0.0,
|
| 82 |
+
"mask_time_length": 10,
|
| 83 |
+
"mask_time_min_masks": 2,
|
| 84 |
+
"mask_time_prob": 0.65,
|
| 85 |
+
"model_type": "wav2vec2",
|
| 86 |
+
"num_adapter_layers": 3,
|
| 87 |
+
"num_attention_heads": 6,
|
| 88 |
+
"num_codevector_groups": 2,
|
| 89 |
+
"num_codevectors_per_group": 320,
|
| 90 |
+
"num_conv_pos_embedding_groups": 16,
|
| 91 |
+
"num_conv_pos_embeddings": 128,
|
| 92 |
+
"num_feat_extract_layers": 7,
|
| 93 |
+
"num_hidden_layers": 6,
|
| 94 |
+
"num_negatives": 100,
|
| 95 |
+
"output_hidden_size": 384,
|
| 96 |
+
"pad_token_id": 0,
|
| 97 |
+
"proj_codevector_dim": 128,
|
| 98 |
+
"tdnn_dilation": [
|
| 99 |
+
1,
|
| 100 |
+
2,
|
| 101 |
+
3,
|
| 102 |
+
1,
|
| 103 |
+
1
|
| 104 |
+
],
|
| 105 |
+
"tdnn_dim": [
|
| 106 |
+
512,
|
| 107 |
+
512,
|
| 108 |
+
512,
|
| 109 |
+
512,
|
| 110 |
+
1500
|
| 111 |
+
],
|
| 112 |
+
"tdnn_kernel": [
|
| 113 |
+
5,
|
| 114 |
+
3,
|
| 115 |
+
3,
|
| 116 |
+
1,
|
| 117 |
+
1
|
| 118 |
+
],
|
| 119 |
+
"torch_dtype": "float32",
|
| 120 |
+
"transformers_version": "4.39.0.dev0",
|
| 121 |
+
"use_weighted_layer_sum": false,
|
| 122 |
+
"vocab_size": 32,
|
| 123 |
+
"xvector_output_dim": 512
|
| 124 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12b72afca351d5838b740f0ec6003c2e1e2a8c0f5156e6629ad3e2ef735bb540
|
| 3 |
+
size 52151348
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd83ae850a6f38e955a9a0ecf3728e21f005733ae6ae7f948544118441a4714b
|
| 3 |
+
size 95909946
|
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": 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 sha256:0524d4c5bb50cb3a888e246202453f6e7f310c8e7d978c2791811f64716d2d2c
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66b1ff4c92709ee514d60150ad1c67a13e001dac71642548c74f443c1156d358
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2793 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7937062937062938,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-5Class-train-test-finetune/checkpoint-4122",
|
| 4 |
+
"epoch": 224.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5432,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.99,
|
| 13 |
+
"eval_accuracy": 0.34265734265734266,
|
| 14 |
+
"eval_loss": 1.5984586477279663,
|
| 15 |
+
"eval_runtime": 5.3437,
|
| 16 |
+
"eval_samples_per_second": 53.521,
|
| 17 |
+
"eval_steps_per_second": 3.368,
|
| 18 |
+
"step": 24
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.98,
|
| 22 |
+
"eval_accuracy": 0.33916083916083917,
|
| 23 |
+
"eval_loss": 1.5969289541244507,
|
| 24 |
+
"eval_runtime": 3.8653,
|
| 25 |
+
"eval_samples_per_second": 73.992,
|
| 26 |
+
"eval_steps_per_second": 4.657,
|
| 27 |
+
"step": 48
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.06,
|
| 31 |
+
"grad_norm": 1.0544973611831665,
|
| 32 |
+
"learning_rate": 2.4999999999999998e-06,
|
| 33 |
+
"loss": 1.5969,
|
| 34 |
+
"step": 50
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.97,
|
| 38 |
+
"eval_accuracy": 0.32867132867132864,
|
| 39 |
+
"eval_loss": 1.5943816900253296,
|
| 40 |
+
"eval_runtime": 6.1748,
|
| 41 |
+
"eval_samples_per_second": 46.317,
|
| 42 |
+
"eval_steps_per_second": 2.915,
|
| 43 |
+
"step": 72
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.3146853146853147,
|
| 48 |
+
"eval_loss": 1.5906767845153809,
|
| 49 |
+
"eval_runtime": 5.1678,
|
| 50 |
+
"eval_samples_per_second": 55.343,
|
| 51 |
+
"eval_steps_per_second": 3.483,
|
| 52 |
+
"step": 97
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.12,
|
| 56 |
+
"grad_norm": 0.8443157076835632,
|
| 57 |
+
"learning_rate": 4.9999999999999996e-06,
|
| 58 |
+
"loss": 1.5896,
|
| 59 |
+
"step": 100
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.99,
|
| 63 |
+
"eval_accuracy": 0.2972027972027972,
|
| 64 |
+
"eval_loss": 1.5860023498535156,
|
| 65 |
+
"eval_runtime": 4.9416,
|
| 66 |
+
"eval_samples_per_second": 57.876,
|
| 67 |
+
"eval_steps_per_second": 3.643,
|
| 68 |
+
"step": 121
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.98,
|
| 72 |
+
"eval_accuracy": 0.2692307692307692,
|
| 73 |
+
"eval_loss": 1.5806005001068115,
|
| 74 |
+
"eval_runtime": 4.1837,
|
| 75 |
+
"eval_samples_per_second": 68.36,
|
| 76 |
+
"eval_steps_per_second": 4.302,
|
| 77 |
+
"step": 145
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.19,
|
| 81 |
+
"grad_norm": 1.0938074588775635,
|
| 82 |
+
"learning_rate": 7.5e-06,
|
| 83 |
+
"loss": 1.5743,
|
| 84 |
+
"step": 150
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.97,
|
| 88 |
+
"eval_accuracy": 0.25874125874125875,
|
| 89 |
+
"eval_loss": 1.5742768049240112,
|
| 90 |
+
"eval_runtime": 7.1914,
|
| 91 |
+
"eval_samples_per_second": 39.77,
|
| 92 |
+
"eval_steps_per_second": 2.503,
|
| 93 |
+
"step": 169
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.0,
|
| 97 |
+
"eval_accuracy": 0.23426573426573427,
|
| 98 |
+
"eval_loss": 1.5664165019989014,
|
| 99 |
+
"eval_runtime": 5.6489,
|
| 100 |
+
"eval_samples_per_second": 50.629,
|
| 101 |
+
"eval_steps_per_second": 3.186,
|
| 102 |
+
"step": 194
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.25,
|
| 106 |
+
"grad_norm": 0.9692079424858093,
|
| 107 |
+
"learning_rate": 9.999999999999999e-06,
|
| 108 |
+
"loss": 1.5508,
|
| 109 |
+
"step": 200
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 8.99,
|
| 113 |
+
"eval_accuracy": 0.22727272727272727,
|
| 114 |
+
"eval_loss": 1.557572841644287,
|
| 115 |
+
"eval_runtime": 5.5182,
|
| 116 |
+
"eval_samples_per_second": 51.828,
|
| 117 |
+
"eval_steps_per_second": 3.262,
|
| 118 |
+
"step": 218
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 9.98,
|
| 122 |
+
"eval_accuracy": 0.22727272727272727,
|
| 123 |
+
"eval_loss": 1.5482373237609863,
|
| 124 |
+
"eval_runtime": 5.3205,
|
| 125 |
+
"eval_samples_per_second": 53.754,
|
| 126 |
+
"eval_steps_per_second": 3.383,
|
| 127 |
+
"step": 242
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 10.31,
|
| 131 |
+
"grad_norm": 1.02046799659729,
|
| 132 |
+
"learning_rate": 1.25e-05,
|
| 133 |
+
"loss": 1.5157,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.97,
|
| 138 |
+
"eval_accuracy": 0.22727272727272727,
|
| 139 |
+
"eval_loss": 1.539355993270874,
|
| 140 |
+
"eval_runtime": 6.3116,
|
| 141 |
+
"eval_samples_per_second": 45.313,
|
| 142 |
+
"eval_steps_per_second": 2.852,
|
| 143 |
+
"step": 266
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 12.0,
|
| 147 |
+
"eval_accuracy": 0.22727272727272727,
|
| 148 |
+
"eval_loss": 1.5350520610809326,
|
| 149 |
+
"eval_runtime": 4.3422,
|
| 150 |
+
"eval_samples_per_second": 65.865,
|
| 151 |
+
"eval_steps_per_second": 4.145,
|
| 152 |
+
"step": 291
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 12.37,
|
| 156 |
+
"grad_norm": 1.6058833599090576,
|
| 157 |
+
"learning_rate": 1.5e-05,
|
| 158 |
+
"loss": 1.4534,
|
| 159 |
+
"step": 300
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 12.99,
|
| 163 |
+
"eval_accuracy": 0.22727272727272727,
|
| 164 |
+
"eval_loss": 1.5525730848312378,
|
| 165 |
+
"eval_runtime": 5.245,
|
| 166 |
+
"eval_samples_per_second": 54.528,
|
| 167 |
+
"eval_steps_per_second": 3.432,
|
| 168 |
+
"step": 315
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 13.98,
|
| 172 |
+
"eval_accuracy": 0.22727272727272727,
|
| 173 |
+
"eval_loss": 1.599926471710205,
|
| 174 |
+
"eval_runtime": 6.0088,
|
| 175 |
+
"eval_samples_per_second": 47.597,
|
| 176 |
+
"eval_steps_per_second": 2.996,
|
| 177 |
+
"step": 339
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 14.43,
|
| 181 |
+
"grad_norm": 0.8243080377578735,
|
| 182 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 183 |
+
"loss": 1.3638,
|
| 184 |
+
"step": 350
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 14.97,
|
| 188 |
+
"eval_accuracy": 0.22727272727272727,
|
| 189 |
+
"eval_loss": 1.5896875858306885,
|
| 190 |
+
"eval_runtime": 4.8752,
|
| 191 |
+
"eval_samples_per_second": 58.664,
|
| 192 |
+
"eval_steps_per_second": 3.692,
|
| 193 |
+
"step": 363
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 16.0,
|
| 197 |
+
"eval_accuracy": 0.26573426573426573,
|
| 198 |
+
"eval_loss": 1.560091495513916,
|
| 199 |
+
"eval_runtime": 5.5082,
|
| 200 |
+
"eval_samples_per_second": 51.922,
|
| 201 |
+
"eval_steps_per_second": 3.268,
|
| 202 |
+
"step": 388
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 16.49,
|
| 206 |
+
"grad_norm": 0.7977257370948792,
|
| 207 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 208 |
+
"loss": 1.2951,
|
| 209 |
+
"step": 400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 16.99,
|
| 213 |
+
"eval_accuracy": 0.2937062937062937,
|
| 214 |
+
"eval_loss": 1.5349317789077759,
|
| 215 |
+
"eval_runtime": 4.7526,
|
| 216 |
+
"eval_samples_per_second": 60.178,
|
| 217 |
+
"eval_steps_per_second": 3.787,
|
| 218 |
+
"step": 412
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 17.98,
|
| 222 |
+
"eval_accuracy": 0.34265734265734266,
|
| 223 |
+
"eval_loss": 1.5053907632827759,
|
| 224 |
+
"eval_runtime": 4.8638,
|
| 225 |
+
"eval_samples_per_second": 58.801,
|
| 226 |
+
"eval_steps_per_second": 3.701,
|
| 227 |
+
"step": 436
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 18.56,
|
| 231 |
+
"grad_norm": 0.7064552903175354,
|
| 232 |
+
"learning_rate": 2.25e-05,
|
| 233 |
+
"loss": 1.2369,
|
| 234 |
+
"step": 450
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 18.97,
|
| 238 |
+
"eval_accuracy": 0.3741258741258741,
|
| 239 |
+
"eval_loss": 1.4689087867736816,
|
| 240 |
+
"eval_runtime": 4.3712,
|
| 241 |
+
"eval_samples_per_second": 65.428,
|
| 242 |
+
"eval_steps_per_second": 4.118,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 20.0,
|
| 247 |
+
"eval_accuracy": 0.4370629370629371,
|
| 248 |
+
"eval_loss": 1.404613971710205,
|
| 249 |
+
"eval_runtime": 4.7203,
|
| 250 |
+
"eval_samples_per_second": 60.59,
|
| 251 |
+
"eval_steps_per_second": 3.813,
|
| 252 |
+
"step": 485
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 20.62,
|
| 256 |
+
"grad_norm": 0.598238468170166,
|
| 257 |
+
"learning_rate": 2.5e-05,
|
| 258 |
+
"loss": 1.1566,
|
| 259 |
+
"step": 500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 20.99,
|
| 263 |
+
"eval_accuracy": 0.4405594405594406,
|
| 264 |
+
"eval_loss": 1.3691043853759766,
|
| 265 |
+
"eval_runtime": 6.6443,
|
| 266 |
+
"eval_samples_per_second": 43.044,
|
| 267 |
+
"eval_steps_per_second": 2.709,
|
| 268 |
+
"step": 509
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 21.98,
|
| 272 |
+
"eval_accuracy": 0.4825174825174825,
|
| 273 |
+
"eval_loss": 1.3120107650756836,
|
| 274 |
+
"eval_runtime": 4.9585,
|
| 275 |
+
"eval_samples_per_second": 57.679,
|
| 276 |
+
"eval_steps_per_second": 3.63,
|
| 277 |
+
"step": 533
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 22.68,
|
| 281 |
+
"grad_norm": 0.682925820350647,
|
| 282 |
+
"learning_rate": 2.75e-05,
|
| 283 |
+
"loss": 1.0676,
|
| 284 |
+
"step": 550
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 22.97,
|
| 288 |
+
"eval_accuracy": 0.486013986013986,
|
| 289 |
+
"eval_loss": 1.2839338779449463,
|
| 290 |
+
"eval_runtime": 4.0382,
|
| 291 |
+
"eval_samples_per_second": 70.824,
|
| 292 |
+
"eval_steps_per_second": 4.457,
|
| 293 |
+
"step": 557
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 24.0,
|
| 297 |
+
"eval_accuracy": 0.5104895104895105,
|
| 298 |
+
"eval_loss": 1.2549891471862793,
|
| 299 |
+
"eval_runtime": 5.1896,
|
| 300 |
+
"eval_samples_per_second": 55.11,
|
| 301 |
+
"eval_steps_per_second": 3.468,
|
| 302 |
+
"step": 582
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 24.74,
|
| 306 |
+
"grad_norm": 1.1368101835250854,
|
| 307 |
+
"learning_rate": 3e-05,
|
| 308 |
+
"loss": 0.992,
|
| 309 |
+
"step": 600
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 24.99,
|
| 313 |
+
"eval_accuracy": 0.5209790209790209,
|
| 314 |
+
"eval_loss": 1.2106566429138184,
|
| 315 |
+
"eval_runtime": 6.8941,
|
| 316 |
+
"eval_samples_per_second": 41.485,
|
| 317 |
+
"eval_steps_per_second": 2.611,
|
| 318 |
+
"step": 606
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 25.98,
|
| 322 |
+
"eval_accuracy": 0.5384615384615384,
|
| 323 |
+
"eval_loss": 1.1711338758468628,
|
| 324 |
+
"eval_runtime": 4.9707,
|
| 325 |
+
"eval_samples_per_second": 57.537,
|
| 326 |
+
"eval_steps_per_second": 3.621,
|
| 327 |
+
"step": 630
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 26.8,
|
| 331 |
+
"grad_norm": 0.9649831056594849,
|
| 332 |
+
"learning_rate": 2.9722222222222223e-05,
|
| 333 |
+
"loss": 0.9272,
|
| 334 |
+
"step": 650
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 26.97,
|
| 338 |
+
"eval_accuracy": 0.5594405594405595,
|
| 339 |
+
"eval_loss": 1.1318116188049316,
|
| 340 |
+
"eval_runtime": 5.5564,
|
| 341 |
+
"eval_samples_per_second": 51.472,
|
| 342 |
+
"eval_steps_per_second": 3.24,
|
| 343 |
+
"step": 654
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 28.0,
|
| 347 |
+
"eval_accuracy": 0.6153846153846154,
|
| 348 |
+
"eval_loss": 1.0594333410263062,
|
| 349 |
+
"eval_runtime": 4.6773,
|
| 350 |
+
"eval_samples_per_second": 61.147,
|
| 351 |
+
"eval_steps_per_second": 3.848,
|
| 352 |
+
"step": 679
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 28.87,
|
| 356 |
+
"grad_norm": 0.883937418460846,
|
| 357 |
+
"learning_rate": 2.9444444444444445e-05,
|
| 358 |
+
"loss": 0.8478,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 28.99,
|
| 363 |
+
"eval_accuracy": 0.6013986013986014,
|
| 364 |
+
"eval_loss": 1.054669737815857,
|
| 365 |
+
"eval_runtime": 4.9219,
|
| 366 |
+
"eval_samples_per_second": 58.108,
|
| 367 |
+
"eval_steps_per_second": 3.657,
|
| 368 |
+
"step": 703
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 29.98,
|
| 372 |
+
"eval_accuracy": 0.6363636363636364,
|
| 373 |
+
"eval_loss": 0.9822685122489929,
|
| 374 |
+
"eval_runtime": 6.3133,
|
| 375 |
+
"eval_samples_per_second": 45.302,
|
| 376 |
+
"eval_steps_per_second": 2.851,
|
| 377 |
+
"step": 727
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 30.93,
|
| 381 |
+
"grad_norm": 1.3742878437042236,
|
| 382 |
+
"learning_rate": 2.9166666666666666e-05,
|
| 383 |
+
"loss": 0.7627,
|
| 384 |
+
"step": 750
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 30.97,
|
| 388 |
+
"eval_accuracy": 0.6398601398601399,
|
| 389 |
+
"eval_loss": 1.00295090675354,
|
| 390 |
+
"eval_runtime": 6.154,
|
| 391 |
+
"eval_samples_per_second": 46.473,
|
| 392 |
+
"eval_steps_per_second": 2.925,
|
| 393 |
+
"step": 751
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 32.0,
|
| 397 |
+
"eval_accuracy": 0.6608391608391608,
|
| 398 |
+
"eval_loss": 0.930969774723053,
|
| 399 |
+
"eval_runtime": 5.6747,
|
| 400 |
+
"eval_samples_per_second": 50.399,
|
| 401 |
+
"eval_steps_per_second": 3.172,
|
| 402 |
+
"step": 776
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 32.99,
|
| 406 |
+
"grad_norm": 1.329268217086792,
|
| 407 |
+
"learning_rate": 2.8888888888888888e-05,
|
| 408 |
+
"loss": 0.7266,
|
| 409 |
+
"step": 800
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 32.99,
|
| 413 |
+
"eval_accuracy": 0.6678321678321678,
|
| 414 |
+
"eval_loss": 0.9228739738464355,
|
| 415 |
+
"eval_runtime": 5.382,
|
| 416 |
+
"eval_samples_per_second": 53.14,
|
| 417 |
+
"eval_steps_per_second": 3.344,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 33.98,
|
| 422 |
+
"eval_accuracy": 0.6958041958041958,
|
| 423 |
+
"eval_loss": 0.8684509992599487,
|
| 424 |
+
"eval_runtime": 4.8497,
|
| 425 |
+
"eval_samples_per_second": 58.973,
|
| 426 |
+
"eval_steps_per_second": 3.712,
|
| 427 |
+
"step": 824
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 34.97,
|
| 431 |
+
"eval_accuracy": 0.6643356643356644,
|
| 432 |
+
"eval_loss": 0.8954732418060303,
|
| 433 |
+
"eval_runtime": 5.2083,
|
| 434 |
+
"eval_samples_per_second": 54.912,
|
| 435 |
+
"eval_steps_per_second": 3.456,
|
| 436 |
+
"step": 848
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 35.05,
|
| 440 |
+
"grad_norm": 1.3892701864242554,
|
| 441 |
+
"learning_rate": 2.8611111111111113e-05,
|
| 442 |
+
"loss": 0.6906,
|
| 443 |
+
"step": 850
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 36.0,
|
| 447 |
+
"eval_accuracy": 0.6713286713286714,
|
| 448 |
+
"eval_loss": 0.9125654101371765,
|
| 449 |
+
"eval_runtime": 5.3068,
|
| 450 |
+
"eval_samples_per_second": 53.894,
|
| 451 |
+
"eval_steps_per_second": 3.392,
|
| 452 |
+
"step": 873
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 36.99,
|
| 456 |
+
"eval_accuracy": 0.6923076923076923,
|
| 457 |
+
"eval_loss": 0.8543534874916077,
|
| 458 |
+
"eval_runtime": 4.3351,
|
| 459 |
+
"eval_samples_per_second": 65.974,
|
| 460 |
+
"eval_steps_per_second": 4.152,
|
| 461 |
+
"step": 897
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 37.11,
|
| 465 |
+
"grad_norm": 0.836291491985321,
|
| 466 |
+
"learning_rate": 2.8333333333333332e-05,
|
| 467 |
+
"loss": 0.6721,
|
| 468 |
+
"step": 900
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 37.98,
|
| 472 |
+
"eval_accuracy": 0.6923076923076923,
|
| 473 |
+
"eval_loss": 0.8480322957038879,
|
| 474 |
+
"eval_runtime": 5.1861,
|
| 475 |
+
"eval_samples_per_second": 55.147,
|
| 476 |
+
"eval_steps_per_second": 3.471,
|
| 477 |
+
"step": 921
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 38.97,
|
| 481 |
+
"eval_accuracy": 0.7097902097902098,
|
| 482 |
+
"eval_loss": 0.8354606628417969,
|
| 483 |
+
"eval_runtime": 6.3247,
|
| 484 |
+
"eval_samples_per_second": 45.22,
|
| 485 |
+
"eval_steps_per_second": 2.846,
|
| 486 |
+
"step": 945
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 39.18,
|
| 490 |
+
"grad_norm": 1.6499431133270264,
|
| 491 |
+
"learning_rate": 2.8055555555555557e-05,
|
| 492 |
+
"loss": 0.6442,
|
| 493 |
+
"step": 950
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 40.0,
|
| 497 |
+
"eval_accuracy": 0.6958041958041958,
|
| 498 |
+
"eval_loss": 0.8412452340126038,
|
| 499 |
+
"eval_runtime": 5.2281,
|
| 500 |
+
"eval_samples_per_second": 54.704,
|
| 501 |
+
"eval_steps_per_second": 3.443,
|
| 502 |
+
"step": 970
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 40.99,
|
| 506 |
+
"eval_accuracy": 0.6888111888111889,
|
| 507 |
+
"eval_loss": 0.8356389999389648,
|
| 508 |
+
"eval_runtime": 4.8326,
|
| 509 |
+
"eval_samples_per_second": 59.181,
|
| 510 |
+
"eval_steps_per_second": 3.725,
|
| 511 |
+
"step": 994
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 41.24,
|
| 515 |
+
"grad_norm": 1.1766818761825562,
|
| 516 |
+
"learning_rate": 2.777777777777778e-05,
|
| 517 |
+
"loss": 0.6465,
|
| 518 |
+
"step": 1000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 41.98,
|
| 522 |
+
"eval_accuracy": 0.7062937062937062,
|
| 523 |
+
"eval_loss": 0.8180016875267029,
|
| 524 |
+
"eval_runtime": 5.7926,
|
| 525 |
+
"eval_samples_per_second": 49.374,
|
| 526 |
+
"eval_steps_per_second": 3.107,
|
| 527 |
+
"step": 1018
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 42.97,
|
| 531 |
+
"eval_accuracy": 0.7027972027972028,
|
| 532 |
+
"eval_loss": 0.8103991150856018,
|
| 533 |
+
"eval_runtime": 5.5185,
|
| 534 |
+
"eval_samples_per_second": 51.825,
|
| 535 |
+
"eval_steps_per_second": 3.262,
|
| 536 |
+
"step": 1042
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 43.3,
|
| 540 |
+
"grad_norm": 0.9722403287887573,
|
| 541 |
+
"learning_rate": 2.75e-05,
|
| 542 |
+
"loss": 0.6086,
|
| 543 |
+
"step": 1050
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 44.0,
|
| 547 |
+
"eval_accuracy": 0.6958041958041958,
|
| 548 |
+
"eval_loss": 0.8162235617637634,
|
| 549 |
+
"eval_runtime": 4.9174,
|
| 550 |
+
"eval_samples_per_second": 58.161,
|
| 551 |
+
"eval_steps_per_second": 3.66,
|
| 552 |
+
"step": 1067
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 44.99,
|
| 556 |
+
"eval_accuracy": 0.7027972027972028,
|
| 557 |
+
"eval_loss": 0.7957289218902588,
|
| 558 |
+
"eval_runtime": 4.6891,
|
| 559 |
+
"eval_samples_per_second": 60.992,
|
| 560 |
+
"eval_steps_per_second": 3.839,
|
| 561 |
+
"step": 1091
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 45.36,
|
| 565 |
+
"grad_norm": 1.269113302230835,
|
| 566 |
+
"learning_rate": 2.7222222222222223e-05,
|
| 567 |
+
"loss": 0.5863,
|
| 568 |
+
"step": 1100
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 45.98,
|
| 572 |
+
"eval_accuracy": 0.6958041958041958,
|
| 573 |
+
"eval_loss": 0.8143528699874878,
|
| 574 |
+
"eval_runtime": 6.6805,
|
| 575 |
+
"eval_samples_per_second": 42.811,
|
| 576 |
+
"eval_steps_per_second": 2.694,
|
| 577 |
+
"step": 1115
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 46.97,
|
| 581 |
+
"eval_accuracy": 0.7027972027972028,
|
| 582 |
+
"eval_loss": 0.78568434715271,
|
| 583 |
+
"eval_runtime": 4.7422,
|
| 584 |
+
"eval_samples_per_second": 60.31,
|
| 585 |
+
"eval_steps_per_second": 3.796,
|
| 586 |
+
"step": 1139
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 47.42,
|
| 590 |
+
"grad_norm": 0.9775255918502808,
|
| 591 |
+
"learning_rate": 2.6944444444444445e-05,
|
| 592 |
+
"loss": 0.5877,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 48.0,
|
| 597 |
+
"eval_accuracy": 0.7132867132867133,
|
| 598 |
+
"eval_loss": 0.7764595150947571,
|
| 599 |
+
"eval_runtime": 5.76,
|
| 600 |
+
"eval_samples_per_second": 49.653,
|
| 601 |
+
"eval_steps_per_second": 3.125,
|
| 602 |
+
"step": 1164
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 48.99,
|
| 606 |
+
"eval_accuracy": 0.6993006993006993,
|
| 607 |
+
"eval_loss": 0.7881478071212769,
|
| 608 |
+
"eval_runtime": 5.4965,
|
| 609 |
+
"eval_samples_per_second": 52.033,
|
| 610 |
+
"eval_steps_per_second": 3.275,
|
| 611 |
+
"step": 1188
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 49.48,
|
| 615 |
+
"grad_norm": 1.540124773979187,
|
| 616 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 617 |
+
"loss": 0.5629,
|
| 618 |
+
"step": 1200
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 49.98,
|
| 622 |
+
"eval_accuracy": 0.7097902097902098,
|
| 623 |
+
"eval_loss": 0.7658265829086304,
|
| 624 |
+
"eval_runtime": 4.731,
|
| 625 |
+
"eval_samples_per_second": 60.452,
|
| 626 |
+
"eval_steps_per_second": 3.805,
|
| 627 |
+
"step": 1212
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 50.97,
|
| 631 |
+
"eval_accuracy": 0.7132867132867133,
|
| 632 |
+
"eval_loss": 0.7723098397254944,
|
| 633 |
+
"eval_runtime": 5.8352,
|
| 634 |
+
"eval_samples_per_second": 49.013,
|
| 635 |
+
"eval_steps_per_second": 3.085,
|
| 636 |
+
"step": 1236
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 51.55,
|
| 640 |
+
"grad_norm": 1.2498500347137451,
|
| 641 |
+
"learning_rate": 2.6388888888888892e-05,
|
| 642 |
+
"loss": 0.5476,
|
| 643 |
+
"step": 1250
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 52.0,
|
| 647 |
+
"eval_accuracy": 0.7097902097902098,
|
| 648 |
+
"eval_loss": 0.7603952884674072,
|
| 649 |
+
"eval_runtime": 4.448,
|
| 650 |
+
"eval_samples_per_second": 64.299,
|
| 651 |
+
"eval_steps_per_second": 4.047,
|
| 652 |
+
"step": 1261
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 52.99,
|
| 656 |
+
"eval_accuracy": 0.7202797202797203,
|
| 657 |
+
"eval_loss": 0.7554137706756592,
|
| 658 |
+
"eval_runtime": 6.4218,
|
| 659 |
+
"eval_samples_per_second": 44.536,
|
| 660 |
+
"eval_steps_per_second": 2.803,
|
| 661 |
+
"step": 1285
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 53.61,
|
| 665 |
+
"grad_norm": 0.9919388890266418,
|
| 666 |
+
"learning_rate": 2.6116666666666667e-05,
|
| 667 |
+
"loss": 0.5357,
|
| 668 |
+
"step": 1300
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 53.98,
|
| 672 |
+
"eval_accuracy": 0.7307692307692307,
|
| 673 |
+
"eval_loss": 0.7458928227424622,
|
| 674 |
+
"eval_runtime": 5.3791,
|
| 675 |
+
"eval_samples_per_second": 53.168,
|
| 676 |
+
"eval_steps_per_second": 3.346,
|
| 677 |
+
"step": 1309
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 54.97,
|
| 681 |
+
"eval_accuracy": 0.7132867132867133,
|
| 682 |
+
"eval_loss": 0.7632877230644226,
|
| 683 |
+
"eval_runtime": 5.278,
|
| 684 |
+
"eval_samples_per_second": 54.187,
|
| 685 |
+
"eval_steps_per_second": 3.41,
|
| 686 |
+
"step": 1333
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 55.67,
|
| 690 |
+
"grad_norm": 1.688183307647705,
|
| 691 |
+
"learning_rate": 2.5838888888888892e-05,
|
| 692 |
+
"loss": 0.5335,
|
| 693 |
+
"step": 1350
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 56.0,
|
| 697 |
+
"eval_accuracy": 0.7167832167832168,
|
| 698 |
+
"eval_loss": 0.768308162689209,
|
| 699 |
+
"eval_runtime": 5.7022,
|
| 700 |
+
"eval_samples_per_second": 50.156,
|
| 701 |
+
"eval_steps_per_second": 3.157,
|
| 702 |
+
"step": 1358
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 56.99,
|
| 706 |
+
"eval_accuracy": 0.7307692307692307,
|
| 707 |
+
"eval_loss": 0.7380541563034058,
|
| 708 |
+
"eval_runtime": 4.522,
|
| 709 |
+
"eval_samples_per_second": 63.247,
|
| 710 |
+
"eval_steps_per_second": 3.981,
|
| 711 |
+
"step": 1382
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 57.73,
|
| 715 |
+
"grad_norm": 1.4895784854888916,
|
| 716 |
+
"learning_rate": 2.556111111111111e-05,
|
| 717 |
+
"loss": 0.5107,
|
| 718 |
+
"step": 1400
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 57.98,
|
| 722 |
+
"eval_accuracy": 0.7377622377622378,
|
| 723 |
+
"eval_loss": 0.7308338284492493,
|
| 724 |
+
"eval_runtime": 4.4787,
|
| 725 |
+
"eval_samples_per_second": 63.857,
|
| 726 |
+
"eval_steps_per_second": 4.019,
|
| 727 |
+
"step": 1406
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 58.97,
|
| 731 |
+
"eval_accuracy": 0.7237762237762237,
|
| 732 |
+
"eval_loss": 0.7441032528877258,
|
| 733 |
+
"eval_runtime": 5.8744,
|
| 734 |
+
"eval_samples_per_second": 48.685,
|
| 735 |
+
"eval_steps_per_second": 3.064,
|
| 736 |
+
"step": 1430
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 59.79,
|
| 740 |
+
"grad_norm": 1.4925004243850708,
|
| 741 |
+
"learning_rate": 2.5283333333333332e-05,
|
| 742 |
+
"loss": 0.5105,
|
| 743 |
+
"step": 1450
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 60.0,
|
| 747 |
+
"eval_accuracy": 0.7307692307692307,
|
| 748 |
+
"eval_loss": 0.7481815218925476,
|
| 749 |
+
"eval_runtime": 7.272,
|
| 750 |
+
"eval_samples_per_second": 39.329,
|
| 751 |
+
"eval_steps_per_second": 2.475,
|
| 752 |
+
"step": 1455
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 60.99,
|
| 756 |
+
"eval_accuracy": 0.7342657342657343,
|
| 757 |
+
"eval_loss": 0.733482301235199,
|
| 758 |
+
"eval_runtime": 4.6235,
|
| 759 |
+
"eval_samples_per_second": 61.858,
|
| 760 |
+
"eval_steps_per_second": 3.893,
|
| 761 |
+
"step": 1479
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 61.86,
|
| 765 |
+
"grad_norm": 1.3200663328170776,
|
| 766 |
+
"learning_rate": 2.5005555555555558e-05,
|
| 767 |
+
"loss": 0.4914,
|
| 768 |
+
"step": 1500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 61.98,
|
| 772 |
+
"eval_accuracy": 0.7447552447552448,
|
| 773 |
+
"eval_loss": 0.7241908311843872,
|
| 774 |
+
"eval_runtime": 4.8198,
|
| 775 |
+
"eval_samples_per_second": 59.338,
|
| 776 |
+
"eval_steps_per_second": 3.735,
|
| 777 |
+
"step": 1503
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 62.97,
|
| 781 |
+
"eval_accuracy": 0.7377622377622378,
|
| 782 |
+
"eval_loss": 0.7321043014526367,
|
| 783 |
+
"eval_runtime": 5.8929,
|
| 784 |
+
"eval_samples_per_second": 48.533,
|
| 785 |
+
"eval_steps_per_second": 3.055,
|
| 786 |
+
"step": 1527
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 63.92,
|
| 790 |
+
"grad_norm": 1.1309747695922852,
|
| 791 |
+
"learning_rate": 2.472777777777778e-05,
|
| 792 |
+
"loss": 0.4839,
|
| 793 |
+
"step": 1550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 64.0,
|
| 797 |
+
"eval_accuracy": 0.7342657342657343,
|
| 798 |
+
"eval_loss": 0.7220665216445923,
|
| 799 |
+
"eval_runtime": 5.8635,
|
| 800 |
+
"eval_samples_per_second": 48.776,
|
| 801 |
+
"eval_steps_per_second": 3.07,
|
| 802 |
+
"step": 1552
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 64.99,
|
| 806 |
+
"eval_accuracy": 0.7412587412587412,
|
| 807 |
+
"eval_loss": 0.7136482000350952,
|
| 808 |
+
"eval_runtime": 4.3102,
|
| 809 |
+
"eval_samples_per_second": 66.354,
|
| 810 |
+
"eval_steps_per_second": 4.176,
|
| 811 |
+
"step": 1576
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 65.98,
|
| 815 |
+
"grad_norm": 1.1314157247543335,
|
| 816 |
+
"learning_rate": 2.4449999999999998e-05,
|
| 817 |
+
"loss": 0.4751,
|
| 818 |
+
"step": 1600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 65.98,
|
| 822 |
+
"eval_accuracy": 0.7412587412587412,
|
| 823 |
+
"eval_loss": 0.7198111414909363,
|
| 824 |
+
"eval_runtime": 4.7841,
|
| 825 |
+
"eval_samples_per_second": 59.781,
|
| 826 |
+
"eval_steps_per_second": 3.762,
|
| 827 |
+
"step": 1600
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 66.97,
|
| 831 |
+
"eval_accuracy": 0.7377622377622378,
|
| 832 |
+
"eval_loss": 0.7145721912384033,
|
| 833 |
+
"eval_runtime": 6.347,
|
| 834 |
+
"eval_samples_per_second": 45.061,
|
| 835 |
+
"eval_steps_per_second": 2.836,
|
| 836 |
+
"step": 1624
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 68.0,
|
| 840 |
+
"eval_accuracy": 0.7447552447552448,
|
| 841 |
+
"eval_loss": 0.6970916390419006,
|
| 842 |
+
"eval_runtime": 5.6871,
|
| 843 |
+
"eval_samples_per_second": 50.289,
|
| 844 |
+
"eval_steps_per_second": 3.165,
|
| 845 |
+
"step": 1649
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 68.04,
|
| 849 |
+
"grad_norm": 2.397585153579712,
|
| 850 |
+
"learning_rate": 2.4172222222222223e-05,
|
| 851 |
+
"loss": 0.4639,
|
| 852 |
+
"step": 1650
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 68.99,
|
| 856 |
+
"eval_accuracy": 0.7272727272727273,
|
| 857 |
+
"eval_loss": 0.7201464176177979,
|
| 858 |
+
"eval_runtime": 4.4157,
|
| 859 |
+
"eval_samples_per_second": 64.769,
|
| 860 |
+
"eval_steps_per_second": 4.076,
|
| 861 |
+
"step": 1673
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 69.98,
|
| 865 |
+
"eval_accuracy": 0.7307692307692307,
|
| 866 |
+
"eval_loss": 0.7244682312011719,
|
| 867 |
+
"eval_runtime": 5.4392,
|
| 868 |
+
"eval_samples_per_second": 52.581,
|
| 869 |
+
"eval_steps_per_second": 3.309,
|
| 870 |
+
"step": 1697
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 70.1,
|
| 874 |
+
"grad_norm": 2.062610387802124,
|
| 875 |
+
"learning_rate": 2.3894444444444445e-05,
|
| 876 |
+
"loss": 0.4581,
|
| 877 |
+
"step": 1700
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 70.97,
|
| 881 |
+
"eval_accuracy": 0.7447552447552448,
|
| 882 |
+
"eval_loss": 0.7077587842941284,
|
| 883 |
+
"eval_runtime": 5.1002,
|
| 884 |
+
"eval_samples_per_second": 56.076,
|
| 885 |
+
"eval_steps_per_second": 3.529,
|
| 886 |
+
"step": 1721
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 72.0,
|
| 890 |
+
"eval_accuracy": 0.7517482517482518,
|
| 891 |
+
"eval_loss": 0.6957913637161255,
|
| 892 |
+
"eval_runtime": 4.4485,
|
| 893 |
+
"eval_samples_per_second": 64.291,
|
| 894 |
+
"eval_steps_per_second": 4.046,
|
| 895 |
+
"step": 1746
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 72.16,
|
| 899 |
+
"grad_norm": 2.7808456420898438,
|
| 900 |
+
"learning_rate": 2.3616666666666667e-05,
|
| 901 |
+
"loss": 0.4643,
|
| 902 |
+
"step": 1750
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 72.99,
|
| 906 |
+
"eval_accuracy": 0.7447552447552448,
|
| 907 |
+
"eval_loss": 0.7036928534507751,
|
| 908 |
+
"eval_runtime": 5.9101,
|
| 909 |
+
"eval_samples_per_second": 48.392,
|
| 910 |
+
"eval_steps_per_second": 3.046,
|
| 911 |
+
"step": 1770
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 73.98,
|
| 915 |
+
"eval_accuracy": 0.7482517482517482,
|
| 916 |
+
"eval_loss": 0.71629399061203,
|
| 917 |
+
"eval_runtime": 6.0211,
|
| 918 |
+
"eval_samples_per_second": 47.5,
|
| 919 |
+
"eval_steps_per_second": 2.989,
|
| 920 |
+
"step": 1794
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 74.23,
|
| 924 |
+
"grad_norm": 1.78495192527771,
|
| 925 |
+
"learning_rate": 2.333888888888889e-05,
|
| 926 |
+
"loss": 0.442,
|
| 927 |
+
"step": 1800
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 74.97,
|
| 931 |
+
"eval_accuracy": 0.7377622377622378,
|
| 932 |
+
"eval_loss": 0.6997957229614258,
|
| 933 |
+
"eval_runtime": 4.4212,
|
| 934 |
+
"eval_samples_per_second": 64.688,
|
| 935 |
+
"eval_steps_per_second": 4.071,
|
| 936 |
+
"step": 1818
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 76.0,
|
| 940 |
+
"eval_accuracy": 0.7447552447552448,
|
| 941 |
+
"eval_loss": 0.6946483850479126,
|
| 942 |
+
"eval_runtime": 4.0507,
|
| 943 |
+
"eval_samples_per_second": 70.605,
|
| 944 |
+
"eval_steps_per_second": 4.444,
|
| 945 |
+
"step": 1843
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 76.29,
|
| 949 |
+
"grad_norm": 1.7383118867874146,
|
| 950 |
+
"learning_rate": 2.306111111111111e-05,
|
| 951 |
+
"loss": 0.4305,
|
| 952 |
+
"step": 1850
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 76.99,
|
| 956 |
+
"eval_accuracy": 0.7552447552447552,
|
| 957 |
+
"eval_loss": 0.6857091784477234,
|
| 958 |
+
"eval_runtime": 4.1718,
|
| 959 |
+
"eval_samples_per_second": 68.556,
|
| 960 |
+
"eval_steps_per_second": 4.315,
|
| 961 |
+
"step": 1867
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 77.98,
|
| 965 |
+
"eval_accuracy": 0.7447552447552448,
|
| 966 |
+
"eval_loss": 0.6936307549476624,
|
| 967 |
+
"eval_runtime": 3.8781,
|
| 968 |
+
"eval_samples_per_second": 73.747,
|
| 969 |
+
"eval_steps_per_second": 4.641,
|
| 970 |
+
"step": 1891
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 78.35,
|
| 974 |
+
"grad_norm": 1.047067403793335,
|
| 975 |
+
"learning_rate": 2.2783333333333336e-05,
|
| 976 |
+
"loss": 0.4416,
|
| 977 |
+
"step": 1900
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 78.97,
|
| 981 |
+
"eval_accuracy": 0.7517482517482518,
|
| 982 |
+
"eval_loss": 0.6965110301971436,
|
| 983 |
+
"eval_runtime": 5.1318,
|
| 984 |
+
"eval_samples_per_second": 55.731,
|
| 985 |
+
"eval_steps_per_second": 3.508,
|
| 986 |
+
"step": 1915
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 80.0,
|
| 990 |
+
"eval_accuracy": 0.7482517482517482,
|
| 991 |
+
"eval_loss": 0.7017127871513367,
|
| 992 |
+
"eval_runtime": 4.3418,
|
| 993 |
+
"eval_samples_per_second": 65.871,
|
| 994 |
+
"eval_steps_per_second": 4.146,
|
| 995 |
+
"step": 1940
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 80.41,
|
| 999 |
+
"grad_norm": 1.5354928970336914,
|
| 1000 |
+
"learning_rate": 2.2505555555555554e-05,
|
| 1001 |
+
"loss": 0.428,
|
| 1002 |
+
"step": 1950
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 80.99,
|
| 1006 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1007 |
+
"eval_loss": 0.6970596313476562,
|
| 1008 |
+
"eval_runtime": 5.973,
|
| 1009 |
+
"eval_samples_per_second": 47.882,
|
| 1010 |
+
"eval_steps_per_second": 3.014,
|
| 1011 |
+
"step": 1964
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 81.98,
|
| 1015 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1016 |
+
"eval_loss": 0.6897542476654053,
|
| 1017 |
+
"eval_runtime": 5.0481,
|
| 1018 |
+
"eval_samples_per_second": 56.655,
|
| 1019 |
+
"eval_steps_per_second": 3.566,
|
| 1020 |
+
"step": 1988
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 82.47,
|
| 1024 |
+
"grad_norm": 1.7141317129135132,
|
| 1025 |
+
"learning_rate": 2.2227777777777776e-05,
|
| 1026 |
+
"loss": 0.4093,
|
| 1027 |
+
"step": 2000
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 82.97,
|
| 1031 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1032 |
+
"eval_loss": 0.7004020810127258,
|
| 1033 |
+
"eval_runtime": 4.1986,
|
| 1034 |
+
"eval_samples_per_second": 68.118,
|
| 1035 |
+
"eval_steps_per_second": 4.287,
|
| 1036 |
+
"step": 2012
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 84.0,
|
| 1040 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1041 |
+
"eval_loss": 0.6867479681968689,
|
| 1042 |
+
"eval_runtime": 4.6871,
|
| 1043 |
+
"eval_samples_per_second": 61.018,
|
| 1044 |
+
"eval_steps_per_second": 3.84,
|
| 1045 |
+
"step": 2037
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 84.54,
|
| 1049 |
+
"grad_norm": 2.0219666957855225,
|
| 1050 |
+
"learning_rate": 2.195e-05,
|
| 1051 |
+
"loss": 0.4148,
|
| 1052 |
+
"step": 2050
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 84.99,
|
| 1056 |
+
"eval_accuracy": 0.7377622377622378,
|
| 1057 |
+
"eval_loss": 0.7070020437240601,
|
| 1058 |
+
"eval_runtime": 5.9326,
|
| 1059 |
+
"eval_samples_per_second": 48.208,
|
| 1060 |
+
"eval_steps_per_second": 3.034,
|
| 1061 |
+
"step": 2061
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 85.98,
|
| 1065 |
+
"eval_accuracy": 0.7447552447552448,
|
| 1066 |
+
"eval_loss": 0.7030305862426758,
|
| 1067 |
+
"eval_runtime": 5.3564,
|
| 1068 |
+
"eval_samples_per_second": 53.394,
|
| 1069 |
+
"eval_steps_per_second": 3.36,
|
| 1070 |
+
"step": 2085
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 86.6,
|
| 1074 |
+
"grad_norm": 1.4678714275360107,
|
| 1075 |
+
"learning_rate": 2.1672222222222223e-05,
|
| 1076 |
+
"loss": 0.3923,
|
| 1077 |
+
"step": 2100
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 86.97,
|
| 1081 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1082 |
+
"eval_loss": 0.678174614906311,
|
| 1083 |
+
"eval_runtime": 3.9745,
|
| 1084 |
+
"eval_samples_per_second": 71.96,
|
| 1085 |
+
"eval_steps_per_second": 4.529,
|
| 1086 |
+
"step": 2109
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 88.0,
|
| 1090 |
+
"eval_accuracy": 0.7412587412587412,
|
| 1091 |
+
"eval_loss": 0.7166118621826172,
|
| 1092 |
+
"eval_runtime": 4.0358,
|
| 1093 |
+
"eval_samples_per_second": 70.866,
|
| 1094 |
+
"eval_steps_per_second": 4.46,
|
| 1095 |
+
"step": 2134
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 88.66,
|
| 1099 |
+
"grad_norm": 1.589543342590332,
|
| 1100 |
+
"learning_rate": 2.1394444444444445e-05,
|
| 1101 |
+
"loss": 0.3964,
|
| 1102 |
+
"step": 2150
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 88.99,
|
| 1106 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1107 |
+
"eval_loss": 0.7075912952423096,
|
| 1108 |
+
"eval_runtime": 5.0331,
|
| 1109 |
+
"eval_samples_per_second": 56.823,
|
| 1110 |
+
"eval_steps_per_second": 3.576,
|
| 1111 |
+
"step": 2158
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 89.98,
|
| 1115 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1116 |
+
"eval_loss": 0.6867172122001648,
|
| 1117 |
+
"eval_runtime": 5.386,
|
| 1118 |
+
"eval_samples_per_second": 53.101,
|
| 1119 |
+
"eval_steps_per_second": 3.342,
|
| 1120 |
+
"step": 2182
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 90.72,
|
| 1124 |
+
"grad_norm": 1.3886605501174927,
|
| 1125 |
+
"learning_rate": 2.1116666666666667e-05,
|
| 1126 |
+
"loss": 0.3846,
|
| 1127 |
+
"step": 2200
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 90.97,
|
| 1131 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1132 |
+
"eval_loss": 0.6913285851478577,
|
| 1133 |
+
"eval_runtime": 5.5324,
|
| 1134 |
+
"eval_samples_per_second": 51.696,
|
| 1135 |
+
"eval_steps_per_second": 3.254,
|
| 1136 |
+
"step": 2206
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 92.0,
|
| 1140 |
+
"eval_accuracy": 0.7482517482517482,
|
| 1141 |
+
"eval_loss": 0.7160294651985168,
|
| 1142 |
+
"eval_runtime": 5.2753,
|
| 1143 |
+
"eval_samples_per_second": 54.215,
|
| 1144 |
+
"eval_steps_per_second": 3.412,
|
| 1145 |
+
"step": 2231
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 92.78,
|
| 1149 |
+
"grad_norm": 2.4106783866882324,
|
| 1150 |
+
"learning_rate": 2.083888888888889e-05,
|
| 1151 |
+
"loss": 0.3654,
|
| 1152 |
+
"step": 2250
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 92.99,
|
| 1156 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1157 |
+
"eval_loss": 0.6765207052230835,
|
| 1158 |
+
"eval_runtime": 5.5671,
|
| 1159 |
+
"eval_samples_per_second": 51.373,
|
| 1160 |
+
"eval_steps_per_second": 3.233,
|
| 1161 |
+
"step": 2255
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 93.98,
|
| 1165 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1166 |
+
"eval_loss": 0.6881967186927795,
|
| 1167 |
+
"eval_runtime": 3.8228,
|
| 1168 |
+
"eval_samples_per_second": 74.814,
|
| 1169 |
+
"eval_steps_per_second": 4.709,
|
| 1170 |
+
"step": 2279
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 94.85,
|
| 1174 |
+
"grad_norm": 0.8871183395385742,
|
| 1175 |
+
"learning_rate": 2.0561111111111114e-05,
|
| 1176 |
+
"loss": 0.3577,
|
| 1177 |
+
"step": 2300
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 94.97,
|
| 1181 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1182 |
+
"eval_loss": 0.6852585673332214,
|
| 1183 |
+
"eval_runtime": 4.7228,
|
| 1184 |
+
"eval_samples_per_second": 60.557,
|
| 1185 |
+
"eval_steps_per_second": 3.811,
|
| 1186 |
+
"step": 2303
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 96.0,
|
| 1190 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1191 |
+
"eval_loss": 0.7158808708190918,
|
| 1192 |
+
"eval_runtime": 5.6504,
|
| 1193 |
+
"eval_samples_per_second": 50.616,
|
| 1194 |
+
"eval_steps_per_second": 3.186,
|
| 1195 |
+
"step": 2328
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 96.91,
|
| 1199 |
+
"grad_norm": 1.0019863843917847,
|
| 1200 |
+
"learning_rate": 2.0283333333333333e-05,
|
| 1201 |
+
"loss": 0.37,
|
| 1202 |
+
"step": 2350
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 96.99,
|
| 1206 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1207 |
+
"eval_loss": 0.6943120360374451,
|
| 1208 |
+
"eval_runtime": 4.8337,
|
| 1209 |
+
"eval_samples_per_second": 59.168,
|
| 1210 |
+
"eval_steps_per_second": 3.724,
|
| 1211 |
+
"step": 2352
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 97.98,
|
| 1215 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1216 |
+
"eval_loss": 0.7010317444801331,
|
| 1217 |
+
"eval_runtime": 4.6874,
|
| 1218 |
+
"eval_samples_per_second": 61.015,
|
| 1219 |
+
"eval_steps_per_second": 3.84,
|
| 1220 |
+
"step": 2376
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 98.97,
|
| 1224 |
+
"grad_norm": 1.2908928394317627,
|
| 1225 |
+
"learning_rate": 2.0005555555555555e-05,
|
| 1226 |
+
"loss": 0.3473,
|
| 1227 |
+
"step": 2400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 98.97,
|
| 1231 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1232 |
+
"eval_loss": 0.693758487701416,
|
| 1233 |
+
"eval_runtime": 4.7585,
|
| 1234 |
+
"eval_samples_per_second": 60.103,
|
| 1235 |
+
"eval_steps_per_second": 3.783,
|
| 1236 |
+
"step": 2400
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 100.0,
|
| 1240 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1241 |
+
"eval_loss": 0.6918778419494629,
|
| 1242 |
+
"eval_runtime": 6.6891,
|
| 1243 |
+
"eval_samples_per_second": 42.756,
|
| 1244 |
+
"eval_steps_per_second": 2.691,
|
| 1245 |
+
"step": 2425
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 100.99,
|
| 1249 |
+
"eval_accuracy": 0.7552447552447552,
|
| 1250 |
+
"eval_loss": 0.6849302053451538,
|
| 1251 |
+
"eval_runtime": 4.4685,
|
| 1252 |
+
"eval_samples_per_second": 64.003,
|
| 1253 |
+
"eval_steps_per_second": 4.028,
|
| 1254 |
+
"step": 2449
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 101.03,
|
| 1258 |
+
"grad_norm": 1.1730871200561523,
|
| 1259 |
+
"learning_rate": 1.972777777777778e-05,
|
| 1260 |
+
"loss": 0.3587,
|
| 1261 |
+
"step": 2450
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 101.98,
|
| 1265 |
+
"eval_accuracy": 0.7587412587412588,
|
| 1266 |
+
"eval_loss": 0.6855939030647278,
|
| 1267 |
+
"eval_runtime": 4.3434,
|
| 1268 |
+
"eval_samples_per_second": 65.847,
|
| 1269 |
+
"eval_steps_per_second": 4.144,
|
| 1270 |
+
"step": 2473
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 102.97,
|
| 1274 |
+
"eval_accuracy": 0.7517482517482518,
|
| 1275 |
+
"eval_loss": 0.7046144604682922,
|
| 1276 |
+
"eval_runtime": 4.7166,
|
| 1277 |
+
"eval_samples_per_second": 60.637,
|
| 1278 |
+
"eval_steps_per_second": 3.816,
|
| 1279 |
+
"step": 2497
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 103.09,
|
| 1283 |
+
"grad_norm": 1.3693217039108276,
|
| 1284 |
+
"learning_rate": 1.945e-05,
|
| 1285 |
+
"loss": 0.3429,
|
| 1286 |
+
"step": 2500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 104.0,
|
| 1290 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1291 |
+
"eval_loss": 0.6892997622489929,
|
| 1292 |
+
"eval_runtime": 5.3868,
|
| 1293 |
+
"eval_samples_per_second": 53.092,
|
| 1294 |
+
"eval_steps_per_second": 3.341,
|
| 1295 |
+
"step": 2522
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 104.99,
|
| 1299 |
+
"eval_accuracy": 0.7622377622377622,
|
| 1300 |
+
"eval_loss": 0.6913393139839172,
|
| 1301 |
+
"eval_runtime": 5.09,
|
| 1302 |
+
"eval_samples_per_second": 56.188,
|
| 1303 |
+
"eval_steps_per_second": 3.536,
|
| 1304 |
+
"step": 2546
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 105.15,
|
| 1308 |
+
"grad_norm": 1.923829436302185,
|
| 1309 |
+
"learning_rate": 1.9172222222222224e-05,
|
| 1310 |
+
"loss": 0.3549,
|
| 1311 |
+
"step": 2550
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 105.98,
|
| 1315 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1316 |
+
"eval_loss": 0.6880810856819153,
|
| 1317 |
+
"eval_runtime": 4.6668,
|
| 1318 |
+
"eval_samples_per_second": 61.283,
|
| 1319 |
+
"eval_steps_per_second": 3.857,
|
| 1320 |
+
"step": 2570
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 106.97,
|
| 1324 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1325 |
+
"eval_loss": 0.7097887396812439,
|
| 1326 |
+
"eval_runtime": 6.4652,
|
| 1327 |
+
"eval_samples_per_second": 44.237,
|
| 1328 |
+
"eval_steps_per_second": 2.784,
|
| 1329 |
+
"step": 2594
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 107.22,
|
| 1333 |
+
"grad_norm": 2.702012062072754,
|
| 1334 |
+
"learning_rate": 1.8894444444444446e-05,
|
| 1335 |
+
"loss": 0.3403,
|
| 1336 |
+
"step": 2600
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 108.0,
|
| 1340 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1341 |
+
"eval_loss": 0.6878336668014526,
|
| 1342 |
+
"eval_runtime": 4.6923,
|
| 1343 |
+
"eval_samples_per_second": 60.951,
|
| 1344 |
+
"eval_steps_per_second": 3.836,
|
| 1345 |
+
"step": 2619
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 108.99,
|
| 1349 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1350 |
+
"eval_loss": 0.695954442024231,
|
| 1351 |
+
"eval_runtime": 4.4809,
|
| 1352 |
+
"eval_samples_per_second": 63.827,
|
| 1353 |
+
"eval_steps_per_second": 4.017,
|
| 1354 |
+
"step": 2643
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 109.28,
|
| 1358 |
+
"grad_norm": 2.3427536487579346,
|
| 1359 |
+
"learning_rate": 1.8616666666666667e-05,
|
| 1360 |
+
"loss": 0.3253,
|
| 1361 |
+
"step": 2650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 109.98,
|
| 1365 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1366 |
+
"eval_loss": 0.7005948424339294,
|
| 1367 |
+
"eval_runtime": 4.8882,
|
| 1368 |
+
"eval_samples_per_second": 58.508,
|
| 1369 |
+
"eval_steps_per_second": 3.682,
|
| 1370 |
+
"step": 2667
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 110.97,
|
| 1374 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1375 |
+
"eval_loss": 0.6916196346282959,
|
| 1376 |
+
"eval_runtime": 5.2891,
|
| 1377 |
+
"eval_samples_per_second": 54.073,
|
| 1378 |
+
"eval_steps_per_second": 3.403,
|
| 1379 |
+
"step": 2691
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 111.34,
|
| 1383 |
+
"grad_norm": 2.178089141845703,
|
| 1384 |
+
"learning_rate": 1.833888888888889e-05,
|
| 1385 |
+
"loss": 0.3332,
|
| 1386 |
+
"step": 2700
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 112.0,
|
| 1390 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1391 |
+
"eval_loss": 0.7059447765350342,
|
| 1392 |
+
"eval_runtime": 4.7437,
|
| 1393 |
+
"eval_samples_per_second": 60.291,
|
| 1394 |
+
"eval_steps_per_second": 3.795,
|
| 1395 |
+
"step": 2716
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 112.99,
|
| 1399 |
+
"eval_accuracy": 0.7867132867132867,
|
| 1400 |
+
"eval_loss": 0.6904045939445496,
|
| 1401 |
+
"eval_runtime": 4.9942,
|
| 1402 |
+
"eval_samples_per_second": 57.267,
|
| 1403 |
+
"eval_steps_per_second": 3.604,
|
| 1404 |
+
"step": 2740
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 113.4,
|
| 1408 |
+
"grad_norm": 1.1625444889068604,
|
| 1409 |
+
"learning_rate": 1.806111111111111e-05,
|
| 1410 |
+
"loss": 0.3188,
|
| 1411 |
+
"step": 2750
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 113.98,
|
| 1415 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1416 |
+
"eval_loss": 0.6970774531364441,
|
| 1417 |
+
"eval_runtime": 6.4809,
|
| 1418 |
+
"eval_samples_per_second": 44.13,
|
| 1419 |
+
"eval_steps_per_second": 2.777,
|
| 1420 |
+
"step": 2764
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 114.97,
|
| 1424 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1425 |
+
"eval_loss": 0.700820803642273,
|
| 1426 |
+
"eval_runtime": 5.2617,
|
| 1427 |
+
"eval_samples_per_second": 54.355,
|
| 1428 |
+
"eval_steps_per_second": 3.421,
|
| 1429 |
+
"step": 2788
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 115.46,
|
| 1433 |
+
"grad_norm": 1.2394715547561646,
|
| 1434 |
+
"learning_rate": 1.7783333333333333e-05,
|
| 1435 |
+
"loss": 0.3112,
|
| 1436 |
+
"step": 2800
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 116.0,
|
| 1440 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1441 |
+
"eval_loss": 0.7002130150794983,
|
| 1442 |
+
"eval_runtime": 5.0937,
|
| 1443 |
+
"eval_samples_per_second": 56.147,
|
| 1444 |
+
"eval_steps_per_second": 3.534,
|
| 1445 |
+
"step": 2813
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 116.99,
|
| 1449 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1450 |
+
"eval_loss": 0.6909505724906921,
|
| 1451 |
+
"eval_runtime": 4.7575,
|
| 1452 |
+
"eval_samples_per_second": 60.116,
|
| 1453 |
+
"eval_steps_per_second": 3.784,
|
| 1454 |
+
"step": 2837
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 117.53,
|
| 1458 |
+
"grad_norm": 2.4334964752197266,
|
| 1459 |
+
"learning_rate": 1.7505555555555558e-05,
|
| 1460 |
+
"loss": 0.3153,
|
| 1461 |
+
"step": 2850
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 117.98,
|
| 1465 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1466 |
+
"eval_loss": 0.6957750916481018,
|
| 1467 |
+
"eval_runtime": 4.8105,
|
| 1468 |
+
"eval_samples_per_second": 59.453,
|
| 1469 |
+
"eval_steps_per_second": 3.742,
|
| 1470 |
+
"step": 2861
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 118.97,
|
| 1474 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1475 |
+
"eval_loss": 0.6867520213127136,
|
| 1476 |
+
"eval_runtime": 4.5411,
|
| 1477 |
+
"eval_samples_per_second": 62.98,
|
| 1478 |
+
"eval_steps_per_second": 3.964,
|
| 1479 |
+
"step": 2885
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 119.59,
|
| 1483 |
+
"grad_norm": 0.769097089767456,
|
| 1484 |
+
"learning_rate": 1.7227777777777777e-05,
|
| 1485 |
+
"loss": 0.3006,
|
| 1486 |
+
"step": 2900
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 120.0,
|
| 1490 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1491 |
+
"eval_loss": 0.6890790462493896,
|
| 1492 |
+
"eval_runtime": 4.5864,
|
| 1493 |
+
"eval_samples_per_second": 62.358,
|
| 1494 |
+
"eval_steps_per_second": 3.925,
|
| 1495 |
+
"step": 2910
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 120.99,
|
| 1499 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1500 |
+
"eval_loss": 0.6889089941978455,
|
| 1501 |
+
"eval_runtime": 6.5804,
|
| 1502 |
+
"eval_samples_per_second": 43.462,
|
| 1503 |
+
"eval_steps_per_second": 2.735,
|
| 1504 |
+
"step": 2934
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 121.65,
|
| 1508 |
+
"grad_norm": 1.8714542388916016,
|
| 1509 |
+
"learning_rate": 1.695e-05,
|
| 1510 |
+
"loss": 0.2967,
|
| 1511 |
+
"step": 2950
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 121.98,
|
| 1515 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1516 |
+
"eval_loss": 0.6935350894927979,
|
| 1517 |
+
"eval_runtime": 4.7491,
|
| 1518 |
+
"eval_samples_per_second": 60.223,
|
| 1519 |
+
"eval_steps_per_second": 3.79,
|
| 1520 |
+
"step": 2958
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 122.97,
|
| 1524 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1525 |
+
"eval_loss": 0.7058219909667969,
|
| 1526 |
+
"eval_runtime": 4.8941,
|
| 1527 |
+
"eval_samples_per_second": 58.438,
|
| 1528 |
+
"eval_steps_per_second": 3.678,
|
| 1529 |
+
"step": 2982
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 123.71,
|
| 1533 |
+
"grad_norm": 2.062924385070801,
|
| 1534 |
+
"learning_rate": 1.6672222222222224e-05,
|
| 1535 |
+
"loss": 0.2939,
|
| 1536 |
+
"step": 3000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 124.0,
|
| 1540 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1541 |
+
"eval_loss": 0.7220865488052368,
|
| 1542 |
+
"eval_runtime": 5.0487,
|
| 1543 |
+
"eval_samples_per_second": 56.648,
|
| 1544 |
+
"eval_steps_per_second": 3.565,
|
| 1545 |
+
"step": 3007
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 124.99,
|
| 1549 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1550 |
+
"eval_loss": 0.6857044696807861,
|
| 1551 |
+
"eval_runtime": 5.6134,
|
| 1552 |
+
"eval_samples_per_second": 50.95,
|
| 1553 |
+
"eval_steps_per_second": 3.207,
|
| 1554 |
+
"step": 3031
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 125.77,
|
| 1558 |
+
"grad_norm": 1.7039302587509155,
|
| 1559 |
+
"learning_rate": 1.6394444444444446e-05,
|
| 1560 |
+
"loss": 0.3101,
|
| 1561 |
+
"step": 3050
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 125.98,
|
| 1565 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1566 |
+
"eval_loss": 0.6742061972618103,
|
| 1567 |
+
"eval_runtime": 5.3609,
|
| 1568 |
+
"eval_samples_per_second": 53.349,
|
| 1569 |
+
"eval_steps_per_second": 3.358,
|
| 1570 |
+
"step": 3055
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 126.97,
|
| 1574 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1575 |
+
"eval_loss": 0.7029407620429993,
|
| 1576 |
+
"eval_runtime": 5.8891,
|
| 1577 |
+
"eval_samples_per_second": 48.564,
|
| 1578 |
+
"eval_steps_per_second": 3.056,
|
| 1579 |
+
"step": 3079
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 127.84,
|
| 1583 |
+
"grad_norm": 1.434970736503601,
|
| 1584 |
+
"learning_rate": 1.6116666666666668e-05,
|
| 1585 |
+
"loss": 0.284,
|
| 1586 |
+
"step": 3100
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 128.0,
|
| 1590 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1591 |
+
"eval_loss": 0.682050347328186,
|
| 1592 |
+
"eval_runtime": 5.1437,
|
| 1593 |
+
"eval_samples_per_second": 55.602,
|
| 1594 |
+
"eval_steps_per_second": 3.499,
|
| 1595 |
+
"step": 3104
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 128.99,
|
| 1599 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1600 |
+
"eval_loss": 0.68370121717453,
|
| 1601 |
+
"eval_runtime": 4.2733,
|
| 1602 |
+
"eval_samples_per_second": 66.927,
|
| 1603 |
+
"eval_steps_per_second": 4.212,
|
| 1604 |
+
"step": 3128
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 129.9,
|
| 1608 |
+
"grad_norm": 1.320789098739624,
|
| 1609 |
+
"learning_rate": 1.583888888888889e-05,
|
| 1610 |
+
"loss": 0.2902,
|
| 1611 |
+
"step": 3150
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 129.98,
|
| 1615 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1616 |
+
"eval_loss": 0.6823462843894958,
|
| 1617 |
+
"eval_runtime": 5.7566,
|
| 1618 |
+
"eval_samples_per_second": 49.682,
|
| 1619 |
+
"eval_steps_per_second": 3.127,
|
| 1620 |
+
"step": 3152
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 130.97,
|
| 1624 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1625 |
+
"eval_loss": 0.6950440406799316,
|
| 1626 |
+
"eval_runtime": 4.9248,
|
| 1627 |
+
"eval_samples_per_second": 58.074,
|
| 1628 |
+
"eval_steps_per_second": 3.655,
|
| 1629 |
+
"step": 3176
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 131.96,
|
| 1633 |
+
"grad_norm": 2.1280930042266846,
|
| 1634 |
+
"learning_rate": 1.556111111111111e-05,
|
| 1635 |
+
"loss": 0.301,
|
| 1636 |
+
"step": 3200
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 132.0,
|
| 1640 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1641 |
+
"eval_loss": 0.6800761818885803,
|
| 1642 |
+
"eval_runtime": 8.1328,
|
| 1643 |
+
"eval_samples_per_second": 35.166,
|
| 1644 |
+
"eval_steps_per_second": 2.213,
|
| 1645 |
+
"step": 3201
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 132.99,
|
| 1649 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1650 |
+
"eval_loss": 0.6867505311965942,
|
| 1651 |
+
"eval_runtime": 4.2532,
|
| 1652 |
+
"eval_samples_per_second": 67.244,
|
| 1653 |
+
"eval_steps_per_second": 4.232,
|
| 1654 |
+
"step": 3225
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 133.98,
|
| 1658 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1659 |
+
"eval_loss": 0.7061284184455872,
|
| 1660 |
+
"eval_runtime": 5.3031,
|
| 1661 |
+
"eval_samples_per_second": 53.93,
|
| 1662 |
+
"eval_steps_per_second": 3.394,
|
| 1663 |
+
"step": 3249
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 134.02,
|
| 1667 |
+
"grad_norm": 1.532638669013977,
|
| 1668 |
+
"learning_rate": 1.5283333333333333e-05,
|
| 1669 |
+
"loss": 0.2736,
|
| 1670 |
+
"step": 3250
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 134.97,
|
| 1674 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1675 |
+
"eval_loss": 0.7114368677139282,
|
| 1676 |
+
"eval_runtime": 4.6536,
|
| 1677 |
+
"eval_samples_per_second": 61.458,
|
| 1678 |
+
"eval_steps_per_second": 3.868,
|
| 1679 |
+
"step": 3273
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 136.0,
|
| 1683 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1684 |
+
"eval_loss": 0.6914551854133606,
|
| 1685 |
+
"eval_runtime": 4.5505,
|
| 1686 |
+
"eval_samples_per_second": 62.851,
|
| 1687 |
+
"eval_steps_per_second": 3.956,
|
| 1688 |
+
"step": 3298
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 136.08,
|
| 1692 |
+
"grad_norm": 2.0108492374420166,
|
| 1693 |
+
"learning_rate": 1.5005555555555555e-05,
|
| 1694 |
+
"loss": 0.2931,
|
| 1695 |
+
"step": 3300
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 136.99,
|
| 1699 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1700 |
+
"eval_loss": 0.7055917978286743,
|
| 1701 |
+
"eval_runtime": 5.3067,
|
| 1702 |
+
"eval_samples_per_second": 53.894,
|
| 1703 |
+
"eval_steps_per_second": 3.392,
|
| 1704 |
+
"step": 3322
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 137.98,
|
| 1708 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1709 |
+
"eval_loss": 0.7026935815811157,
|
| 1710 |
+
"eval_runtime": 5.186,
|
| 1711 |
+
"eval_samples_per_second": 55.149,
|
| 1712 |
+
"eval_steps_per_second": 3.471,
|
| 1713 |
+
"step": 3346
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 138.14,
|
| 1717 |
+
"grad_norm": 1.0804469585418701,
|
| 1718 |
+
"learning_rate": 1.4727777777777779e-05,
|
| 1719 |
+
"loss": 0.2864,
|
| 1720 |
+
"step": 3350
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 138.97,
|
| 1724 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1725 |
+
"eval_loss": 0.6983500719070435,
|
| 1726 |
+
"eval_runtime": 6.955,
|
| 1727 |
+
"eval_samples_per_second": 41.122,
|
| 1728 |
+
"eval_steps_per_second": 2.588,
|
| 1729 |
+
"step": 3370
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 140.0,
|
| 1733 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1734 |
+
"eval_loss": 0.7168787121772766,
|
| 1735 |
+
"eval_runtime": 4.234,
|
| 1736 |
+
"eval_samples_per_second": 67.548,
|
| 1737 |
+
"eval_steps_per_second": 4.251,
|
| 1738 |
+
"step": 3395
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 140.21,
|
| 1742 |
+
"grad_norm": 2.370694637298584,
|
| 1743 |
+
"learning_rate": 1.445e-05,
|
| 1744 |
+
"loss": 0.2765,
|
| 1745 |
+
"step": 3400
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 140.99,
|
| 1749 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1750 |
+
"eval_loss": 0.6960318088531494,
|
| 1751 |
+
"eval_runtime": 5.0294,
|
| 1752 |
+
"eval_samples_per_second": 56.865,
|
| 1753 |
+
"eval_steps_per_second": 3.579,
|
| 1754 |
+
"step": 3419
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 141.98,
|
| 1758 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1759 |
+
"eval_loss": 0.6990492343902588,
|
| 1760 |
+
"eval_runtime": 5.2727,
|
| 1761 |
+
"eval_samples_per_second": 54.242,
|
| 1762 |
+
"eval_steps_per_second": 3.414,
|
| 1763 |
+
"step": 3443
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 142.27,
|
| 1767 |
+
"grad_norm": 1.6676194667816162,
|
| 1768 |
+
"learning_rate": 1.4172222222222222e-05,
|
| 1769 |
+
"loss": 0.2808,
|
| 1770 |
+
"step": 3450
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 142.97,
|
| 1774 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1775 |
+
"eval_loss": 0.706200897693634,
|
| 1776 |
+
"eval_runtime": 4.5273,
|
| 1777 |
+
"eval_samples_per_second": 63.173,
|
| 1778 |
+
"eval_steps_per_second": 3.976,
|
| 1779 |
+
"step": 3467
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 144.0,
|
| 1783 |
+
"eval_accuracy": 0.7657342657342657,
|
| 1784 |
+
"eval_loss": 0.6821764707565308,
|
| 1785 |
+
"eval_runtime": 5.3614,
|
| 1786 |
+
"eval_samples_per_second": 53.344,
|
| 1787 |
+
"eval_steps_per_second": 3.357,
|
| 1788 |
+
"step": 3492
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 144.33,
|
| 1792 |
+
"grad_norm": 1.9151145219802856,
|
| 1793 |
+
"learning_rate": 1.3894444444444444e-05,
|
| 1794 |
+
"loss": 0.2712,
|
| 1795 |
+
"step": 3500
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 144.99,
|
| 1799 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1800 |
+
"eval_loss": 0.7063603401184082,
|
| 1801 |
+
"eval_runtime": 4.9088,
|
| 1802 |
+
"eval_samples_per_second": 58.263,
|
| 1803 |
+
"eval_steps_per_second": 3.667,
|
| 1804 |
+
"step": 3516
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 145.98,
|
| 1808 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1809 |
+
"eval_loss": 0.7150112390518188,
|
| 1810 |
+
"eval_runtime": 7.2044,
|
| 1811 |
+
"eval_samples_per_second": 39.698,
|
| 1812 |
+
"eval_steps_per_second": 2.498,
|
| 1813 |
+
"step": 3540
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 146.39,
|
| 1817 |
+
"grad_norm": 1.5093848705291748,
|
| 1818 |
+
"learning_rate": 1.3622222222222223e-05,
|
| 1819 |
+
"loss": 0.2726,
|
| 1820 |
+
"step": 3550
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 146.97,
|
| 1824 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1825 |
+
"eval_loss": 0.696849524974823,
|
| 1826 |
+
"eval_runtime": 4.9386,
|
| 1827 |
+
"eval_samples_per_second": 57.911,
|
| 1828 |
+
"eval_steps_per_second": 3.645,
|
| 1829 |
+
"step": 3564
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 148.0,
|
| 1833 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1834 |
+
"eval_loss": 0.7086759209632874,
|
| 1835 |
+
"eval_runtime": 4.4363,
|
| 1836 |
+
"eval_samples_per_second": 64.468,
|
| 1837 |
+
"eval_steps_per_second": 4.057,
|
| 1838 |
+
"step": 3589
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 148.45,
|
| 1842 |
+
"grad_norm": 1.4403679370880127,
|
| 1843 |
+
"learning_rate": 1.3344444444444444e-05,
|
| 1844 |
+
"loss": 0.2607,
|
| 1845 |
+
"step": 3600
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 148.99,
|
| 1849 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1850 |
+
"eval_loss": 0.7129560112953186,
|
| 1851 |
+
"eval_runtime": 5.3809,
|
| 1852 |
+
"eval_samples_per_second": 53.15,
|
| 1853 |
+
"eval_steps_per_second": 3.345,
|
| 1854 |
+
"step": 3613
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 149.98,
|
| 1858 |
+
"eval_accuracy": 0.7902097902097902,
|
| 1859 |
+
"eval_loss": 0.7080287933349609,
|
| 1860 |
+
"eval_runtime": 5.8187,
|
| 1861 |
+
"eval_samples_per_second": 49.152,
|
| 1862 |
+
"eval_steps_per_second": 3.093,
|
| 1863 |
+
"step": 3637
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 150.52,
|
| 1867 |
+
"grad_norm": 2.036515235900879,
|
| 1868 |
+
"learning_rate": 1.3066666666666666e-05,
|
| 1869 |
+
"loss": 0.2546,
|
| 1870 |
+
"step": 3650
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 150.97,
|
| 1874 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1875 |
+
"eval_loss": 0.7088435888290405,
|
| 1876 |
+
"eval_runtime": 4.8742,
|
| 1877 |
+
"eval_samples_per_second": 58.677,
|
| 1878 |
+
"eval_steps_per_second": 3.693,
|
| 1879 |
+
"step": 3661
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 152.0,
|
| 1883 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1884 |
+
"eval_loss": 0.7030193209648132,
|
| 1885 |
+
"eval_runtime": 4.9492,
|
| 1886 |
+
"eval_samples_per_second": 57.787,
|
| 1887 |
+
"eval_steps_per_second": 3.637,
|
| 1888 |
+
"step": 3686
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 152.58,
|
| 1892 |
+
"grad_norm": 1.200052261352539,
|
| 1893 |
+
"learning_rate": 1.2788888888888888e-05,
|
| 1894 |
+
"loss": 0.2563,
|
| 1895 |
+
"step": 3700
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 152.99,
|
| 1899 |
+
"eval_accuracy": 0.7692307692307693,
|
| 1900 |
+
"eval_loss": 0.7077969908714294,
|
| 1901 |
+
"eval_runtime": 4.614,
|
| 1902 |
+
"eval_samples_per_second": 61.985,
|
| 1903 |
+
"eval_steps_per_second": 3.901,
|
| 1904 |
+
"step": 3710
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 153.98,
|
| 1908 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1909 |
+
"eval_loss": 0.700455904006958,
|
| 1910 |
+
"eval_runtime": 5.7657,
|
| 1911 |
+
"eval_samples_per_second": 49.604,
|
| 1912 |
+
"eval_steps_per_second": 3.122,
|
| 1913 |
+
"step": 3734
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 154.64,
|
| 1917 |
+
"grad_norm": 2.2751214504241943,
|
| 1918 |
+
"learning_rate": 1.2511111111111112e-05,
|
| 1919 |
+
"loss": 0.2531,
|
| 1920 |
+
"step": 3750
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 154.97,
|
| 1924 |
+
"eval_accuracy": 0.7727272727272727,
|
| 1925 |
+
"eval_loss": 0.7160292267799377,
|
| 1926 |
+
"eval_runtime": 5.1079,
|
| 1927 |
+
"eval_samples_per_second": 55.992,
|
| 1928 |
+
"eval_steps_per_second": 3.524,
|
| 1929 |
+
"step": 3758
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 156.0,
|
| 1933 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1934 |
+
"eval_loss": 0.7175909876823425,
|
| 1935 |
+
"eval_runtime": 5.4035,
|
| 1936 |
+
"eval_samples_per_second": 52.929,
|
| 1937 |
+
"eval_steps_per_second": 3.331,
|
| 1938 |
+
"step": 3783
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 156.7,
|
| 1942 |
+
"grad_norm": 1.9024412631988525,
|
| 1943 |
+
"learning_rate": 1.2233333333333334e-05,
|
| 1944 |
+
"loss": 0.2446,
|
| 1945 |
+
"step": 3800
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 156.99,
|
| 1949 |
+
"eval_accuracy": 0.7762237762237763,
|
| 1950 |
+
"eval_loss": 0.7190600037574768,
|
| 1951 |
+
"eval_runtime": 4.3633,
|
| 1952 |
+
"eval_samples_per_second": 65.546,
|
| 1953 |
+
"eval_steps_per_second": 4.125,
|
| 1954 |
+
"step": 3807
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 157.98,
|
| 1958 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1959 |
+
"eval_loss": 0.719641387462616,
|
| 1960 |
+
"eval_runtime": 5.0426,
|
| 1961 |
+
"eval_samples_per_second": 56.717,
|
| 1962 |
+
"eval_steps_per_second": 3.57,
|
| 1963 |
+
"step": 3831
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 158.76,
|
| 1967 |
+
"grad_norm": 3.471806287765503,
|
| 1968 |
+
"learning_rate": 1.1955555555555556e-05,
|
| 1969 |
+
"loss": 0.2479,
|
| 1970 |
+
"step": 3850
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 158.97,
|
| 1974 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1975 |
+
"eval_loss": 0.7073430418968201,
|
| 1976 |
+
"eval_runtime": 3.6336,
|
| 1977 |
+
"eval_samples_per_second": 78.711,
|
| 1978 |
+
"eval_steps_per_second": 4.954,
|
| 1979 |
+
"step": 3855
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 160.0,
|
| 1983 |
+
"eval_accuracy": 0.7797202797202797,
|
| 1984 |
+
"eval_loss": 0.7328661680221558,
|
| 1985 |
+
"eval_runtime": 5.2625,
|
| 1986 |
+
"eval_samples_per_second": 54.347,
|
| 1987 |
+
"eval_steps_per_second": 3.42,
|
| 1988 |
+
"step": 3880
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 160.82,
|
| 1992 |
+
"grad_norm": 2.1171793937683105,
|
| 1993 |
+
"learning_rate": 1.1677777777777777e-05,
|
| 1994 |
+
"loss": 0.2523,
|
| 1995 |
+
"step": 3900
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 160.99,
|
| 1999 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2000 |
+
"eval_loss": 0.7158821821212769,
|
| 2001 |
+
"eval_runtime": 6.5877,
|
| 2002 |
+
"eval_samples_per_second": 43.414,
|
| 2003 |
+
"eval_steps_per_second": 2.732,
|
| 2004 |
+
"step": 3904
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 161.98,
|
| 2008 |
+
"eval_accuracy": 0.7692307692307693,
|
| 2009 |
+
"eval_loss": 0.719171404838562,
|
| 2010 |
+
"eval_runtime": 4.5674,
|
| 2011 |
+
"eval_samples_per_second": 62.618,
|
| 2012 |
+
"eval_steps_per_second": 3.941,
|
| 2013 |
+
"step": 3928
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 162.89,
|
| 2017 |
+
"grad_norm": 1.7515395879745483,
|
| 2018 |
+
"learning_rate": 1.1400000000000001e-05,
|
| 2019 |
+
"loss": 0.2523,
|
| 2020 |
+
"step": 3950
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 162.97,
|
| 2024 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2025 |
+
"eval_loss": 0.7281435132026672,
|
| 2026 |
+
"eval_runtime": 4.4866,
|
| 2027 |
+
"eval_samples_per_second": 63.746,
|
| 2028 |
+
"eval_steps_per_second": 4.012,
|
| 2029 |
+
"step": 3952
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 164.0,
|
| 2033 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2034 |
+
"eval_loss": 0.7078841328620911,
|
| 2035 |
+
"eval_runtime": 4.4241,
|
| 2036 |
+
"eval_samples_per_second": 64.645,
|
| 2037 |
+
"eval_steps_per_second": 4.069,
|
| 2038 |
+
"step": 3977
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 164.95,
|
| 2042 |
+
"grad_norm": 1.456335186958313,
|
| 2043 |
+
"learning_rate": 1.1122222222222223e-05,
|
| 2044 |
+
"loss": 0.2422,
|
| 2045 |
+
"step": 4000
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 164.99,
|
| 2049 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2050 |
+
"eval_loss": 0.7161521911621094,
|
| 2051 |
+
"eval_runtime": 5.1239,
|
| 2052 |
+
"eval_samples_per_second": 55.817,
|
| 2053 |
+
"eval_steps_per_second": 3.513,
|
| 2054 |
+
"step": 4001
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 165.98,
|
| 2058 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2059 |
+
"eval_loss": 0.7190020084381104,
|
| 2060 |
+
"eval_runtime": 3.4488,
|
| 2061 |
+
"eval_samples_per_second": 82.926,
|
| 2062 |
+
"eval_steps_per_second": 5.219,
|
| 2063 |
+
"step": 4025
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 166.97,
|
| 2067 |
+
"eval_accuracy": 0.7762237762237763,
|
| 2068 |
+
"eval_loss": 0.7311248779296875,
|
| 2069 |
+
"eval_runtime": 5.0389,
|
| 2070 |
+
"eval_samples_per_second": 56.759,
|
| 2071 |
+
"eval_steps_per_second": 3.572,
|
| 2072 |
+
"step": 4049
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 167.01,
|
| 2076 |
+
"grad_norm": 1.2554075717926025,
|
| 2077 |
+
"learning_rate": 1.0844444444444445e-05,
|
| 2078 |
+
"loss": 0.242,
|
| 2079 |
+
"step": 4050
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 168.0,
|
| 2083 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2084 |
+
"eval_loss": 0.7110462188720703,
|
| 2085 |
+
"eval_runtime": 4.4612,
|
| 2086 |
+
"eval_samples_per_second": 64.108,
|
| 2087 |
+
"eval_steps_per_second": 4.035,
|
| 2088 |
+
"step": 4074
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 168.99,
|
| 2092 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2093 |
+
"eval_loss": 0.7028501629829407,
|
| 2094 |
+
"eval_runtime": 6.955,
|
| 2095 |
+
"eval_samples_per_second": 41.122,
|
| 2096 |
+
"eval_steps_per_second": 2.588,
|
| 2097 |
+
"step": 4098
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 169.07,
|
| 2101 |
+
"grad_norm": 2.8003265857696533,
|
| 2102 |
+
"learning_rate": 1.0566666666666667e-05,
|
| 2103 |
+
"loss": 0.2392,
|
| 2104 |
+
"step": 4100
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 169.98,
|
| 2108 |
+
"eval_accuracy": 0.7937062937062938,
|
| 2109 |
+
"eval_loss": 0.7108554840087891,
|
| 2110 |
+
"eval_runtime": 5.0033,
|
| 2111 |
+
"eval_samples_per_second": 57.162,
|
| 2112 |
+
"eval_steps_per_second": 3.598,
|
| 2113 |
+
"step": 4122
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 170.97,
|
| 2117 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2118 |
+
"eval_loss": 0.7106384634971619,
|
| 2119 |
+
"eval_runtime": 5.1984,
|
| 2120 |
+
"eval_samples_per_second": 55.017,
|
| 2121 |
+
"eval_steps_per_second": 3.463,
|
| 2122 |
+
"step": 4146
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 171.13,
|
| 2126 |
+
"grad_norm": 2.1897969245910645,
|
| 2127 |
+
"learning_rate": 1.028888888888889e-05,
|
| 2128 |
+
"loss": 0.247,
|
| 2129 |
+
"step": 4150
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 172.0,
|
| 2133 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2134 |
+
"eval_loss": 0.7151694297790527,
|
| 2135 |
+
"eval_runtime": 5.1963,
|
| 2136 |
+
"eval_samples_per_second": 55.039,
|
| 2137 |
+
"eval_steps_per_second": 3.464,
|
| 2138 |
+
"step": 4171
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 172.99,
|
| 2142 |
+
"eval_accuracy": 0.7657342657342657,
|
| 2143 |
+
"eval_loss": 0.7254167795181274,
|
| 2144 |
+
"eval_runtime": 4.4466,
|
| 2145 |
+
"eval_samples_per_second": 64.319,
|
| 2146 |
+
"eval_steps_per_second": 4.048,
|
| 2147 |
+
"step": 4195
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 173.2,
|
| 2151 |
+
"grad_norm": 2.769357681274414,
|
| 2152 |
+
"learning_rate": 1.0011111111111112e-05,
|
| 2153 |
+
"loss": 0.2341,
|
| 2154 |
+
"step": 4200
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 173.98,
|
| 2158 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2159 |
+
"eval_loss": 0.7290962338447571,
|
| 2160 |
+
"eval_runtime": 6.2221,
|
| 2161 |
+
"eval_samples_per_second": 45.965,
|
| 2162 |
+
"eval_steps_per_second": 2.893,
|
| 2163 |
+
"step": 4219
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 174.97,
|
| 2167 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2168 |
+
"eval_loss": 0.7088623046875,
|
| 2169 |
+
"eval_runtime": 4.3709,
|
| 2170 |
+
"eval_samples_per_second": 65.433,
|
| 2171 |
+
"eval_steps_per_second": 4.118,
|
| 2172 |
+
"step": 4243
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 175.26,
|
| 2176 |
+
"grad_norm": 2.044703483581543,
|
| 2177 |
+
"learning_rate": 9.733333333333332e-06,
|
| 2178 |
+
"loss": 0.2317,
|
| 2179 |
+
"step": 4250
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 176.0,
|
| 2183 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2184 |
+
"eval_loss": 0.7185826897621155,
|
| 2185 |
+
"eval_runtime": 5.4095,
|
| 2186 |
+
"eval_samples_per_second": 52.87,
|
| 2187 |
+
"eval_steps_per_second": 3.327,
|
| 2188 |
+
"step": 4268
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 176.99,
|
| 2192 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2193 |
+
"eval_loss": 0.7167823314666748,
|
| 2194 |
+
"eval_runtime": 4.9506,
|
| 2195 |
+
"eval_samples_per_second": 57.77,
|
| 2196 |
+
"eval_steps_per_second": 3.636,
|
| 2197 |
+
"step": 4292
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"epoch": 177.32,
|
| 2201 |
+
"grad_norm": 1.078834056854248,
|
| 2202 |
+
"learning_rate": 9.455555555555556e-06,
|
| 2203 |
+
"loss": 0.2269,
|
| 2204 |
+
"step": 4300
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 177.98,
|
| 2208 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2209 |
+
"eval_loss": 0.7237738966941833,
|
| 2210 |
+
"eval_runtime": 4.781,
|
| 2211 |
+
"eval_samples_per_second": 59.82,
|
| 2212 |
+
"eval_steps_per_second": 3.765,
|
| 2213 |
+
"step": 4316
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 178.97,
|
| 2217 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2218 |
+
"eval_loss": 0.7131801247596741,
|
| 2219 |
+
"eval_runtime": 4.6869,
|
| 2220 |
+
"eval_samples_per_second": 61.022,
|
| 2221 |
+
"eval_steps_per_second": 3.841,
|
| 2222 |
+
"step": 4340
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 179.38,
|
| 2226 |
+
"grad_norm": 2.008120536804199,
|
| 2227 |
+
"learning_rate": 9.177777777777778e-06,
|
| 2228 |
+
"loss": 0.2283,
|
| 2229 |
+
"step": 4350
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 180.0,
|
| 2233 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2234 |
+
"eval_loss": 0.7384253144264221,
|
| 2235 |
+
"eval_runtime": 4.5879,
|
| 2236 |
+
"eval_samples_per_second": 62.338,
|
| 2237 |
+
"eval_steps_per_second": 3.923,
|
| 2238 |
+
"step": 4365
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 180.99,
|
| 2242 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2243 |
+
"eval_loss": 0.7002861499786377,
|
| 2244 |
+
"eval_runtime": 5.3238,
|
| 2245 |
+
"eval_samples_per_second": 53.721,
|
| 2246 |
+
"eval_steps_per_second": 3.381,
|
| 2247 |
+
"step": 4389
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"epoch": 181.44,
|
| 2251 |
+
"grad_norm": 1.9518792629241943,
|
| 2252 |
+
"learning_rate": 8.900000000000001e-06,
|
| 2253 |
+
"loss": 0.2303,
|
| 2254 |
+
"step": 4400
|
| 2255 |
+
},
|
| 2256 |
+
{
|
| 2257 |
+
"epoch": 181.98,
|
| 2258 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2259 |
+
"eval_loss": 0.7278482913970947,
|
| 2260 |
+
"eval_runtime": 5.8358,
|
| 2261 |
+
"eval_samples_per_second": 49.008,
|
| 2262 |
+
"eval_steps_per_second": 3.084,
|
| 2263 |
+
"step": 4413
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 182.97,
|
| 2267 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2268 |
+
"eval_loss": 0.7143127918243408,
|
| 2269 |
+
"eval_runtime": 6.1229,
|
| 2270 |
+
"eval_samples_per_second": 46.71,
|
| 2271 |
+
"eval_steps_per_second": 2.94,
|
| 2272 |
+
"step": 4437
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 183.51,
|
| 2276 |
+
"grad_norm": 1.0936890840530396,
|
| 2277 |
+
"learning_rate": 8.622222222222221e-06,
|
| 2278 |
+
"loss": 0.2109,
|
| 2279 |
+
"step": 4450
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 184.0,
|
| 2283 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2284 |
+
"eval_loss": 0.7406834363937378,
|
| 2285 |
+
"eval_runtime": 5.0467,
|
| 2286 |
+
"eval_samples_per_second": 56.671,
|
| 2287 |
+
"eval_steps_per_second": 3.567,
|
| 2288 |
+
"step": 4462
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 184.99,
|
| 2292 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2293 |
+
"eval_loss": 0.7053534388542175,
|
| 2294 |
+
"eval_runtime": 5.279,
|
| 2295 |
+
"eval_samples_per_second": 54.177,
|
| 2296 |
+
"eval_steps_per_second": 3.41,
|
| 2297 |
+
"step": 4486
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 185.57,
|
| 2301 |
+
"grad_norm": 2.9350059032440186,
|
| 2302 |
+
"learning_rate": 8.344444444444445e-06,
|
| 2303 |
+
"loss": 0.2261,
|
| 2304 |
+
"step": 4500
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 185.98,
|
| 2308 |
+
"eval_accuracy": 0.7727272727272727,
|
| 2309 |
+
"eval_loss": 0.7260809540748596,
|
| 2310 |
+
"eval_runtime": 5.4165,
|
| 2311 |
+
"eval_samples_per_second": 52.802,
|
| 2312 |
+
"eval_steps_per_second": 3.323,
|
| 2313 |
+
"step": 4510
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 186.97,
|
| 2317 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2318 |
+
"eval_loss": 0.7240064144134521,
|
| 2319 |
+
"eval_runtime": 5.4866,
|
| 2320 |
+
"eval_samples_per_second": 52.127,
|
| 2321 |
+
"eval_steps_per_second": 3.281,
|
| 2322 |
+
"step": 4534
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 187.63,
|
| 2326 |
+
"grad_norm": 1.8322782516479492,
|
| 2327 |
+
"learning_rate": 8.066666666666667e-06,
|
| 2328 |
+
"loss": 0.2282,
|
| 2329 |
+
"step": 4550
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 188.0,
|
| 2333 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2334 |
+
"eval_loss": 0.7199599146842957,
|
| 2335 |
+
"eval_runtime": 4.6736,
|
| 2336 |
+
"eval_samples_per_second": 61.195,
|
| 2337 |
+
"eval_steps_per_second": 3.851,
|
| 2338 |
+
"step": 4559
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 188.99,
|
| 2342 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2343 |
+
"eval_loss": 0.7102844715118408,
|
| 2344 |
+
"eval_runtime": 5.4219,
|
| 2345 |
+
"eval_samples_per_second": 52.749,
|
| 2346 |
+
"eval_steps_per_second": 3.32,
|
| 2347 |
+
"step": 4583
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 189.69,
|
| 2351 |
+
"grad_norm": 1.8777916431427002,
|
| 2352 |
+
"learning_rate": 7.78888888888889e-06,
|
| 2353 |
+
"loss": 0.2321,
|
| 2354 |
+
"step": 4600
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 189.98,
|
| 2358 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2359 |
+
"eval_loss": 0.7083376049995422,
|
| 2360 |
+
"eval_runtime": 5.9634,
|
| 2361 |
+
"eval_samples_per_second": 47.959,
|
| 2362 |
+
"eval_steps_per_second": 3.018,
|
| 2363 |
+
"step": 4607
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 190.97,
|
| 2367 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2368 |
+
"eval_loss": 0.7244677543640137,
|
| 2369 |
+
"eval_runtime": 5.2078,
|
| 2370 |
+
"eval_samples_per_second": 54.918,
|
| 2371 |
+
"eval_steps_per_second": 3.456,
|
| 2372 |
+
"step": 4631
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 191.75,
|
| 2376 |
+
"grad_norm": 1.5277408361434937,
|
| 2377 |
+
"learning_rate": 7.5111111111111105e-06,
|
| 2378 |
+
"loss": 0.2261,
|
| 2379 |
+
"step": 4650
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 192.0,
|
| 2383 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2384 |
+
"eval_loss": 0.7124583721160889,
|
| 2385 |
+
"eval_runtime": 5.7079,
|
| 2386 |
+
"eval_samples_per_second": 50.106,
|
| 2387 |
+
"eval_steps_per_second": 3.154,
|
| 2388 |
+
"step": 4656
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 192.99,
|
| 2392 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2393 |
+
"eval_loss": 0.7308976054191589,
|
| 2394 |
+
"eval_runtime": 5.3404,
|
| 2395 |
+
"eval_samples_per_second": 53.554,
|
| 2396 |
+
"eval_steps_per_second": 3.371,
|
| 2397 |
+
"step": 4680
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 193.81,
|
| 2401 |
+
"grad_norm": 2.095749616622925,
|
| 2402 |
+
"learning_rate": 7.233333333333333e-06,
|
| 2403 |
+
"loss": 0.2231,
|
| 2404 |
+
"step": 4700
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 193.98,
|
| 2408 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2409 |
+
"eval_loss": 0.7237818837165833,
|
| 2410 |
+
"eval_runtime": 4.6666,
|
| 2411 |
+
"eval_samples_per_second": 61.286,
|
| 2412 |
+
"eval_steps_per_second": 3.857,
|
| 2413 |
+
"step": 4704
|
| 2414 |
+
},
|
| 2415 |
+
{
|
| 2416 |
+
"epoch": 194.97,
|
| 2417 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2418 |
+
"eval_loss": 0.7253320217132568,
|
| 2419 |
+
"eval_runtime": 5.8059,
|
| 2420 |
+
"eval_samples_per_second": 49.261,
|
| 2421 |
+
"eval_steps_per_second": 3.1,
|
| 2422 |
+
"step": 4728
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 195.88,
|
| 2426 |
+
"grad_norm": 1.6955636739730835,
|
| 2427 |
+
"learning_rate": 6.955555555555556e-06,
|
| 2428 |
+
"loss": 0.2083,
|
| 2429 |
+
"step": 4750
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 196.0,
|
| 2433 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2434 |
+
"eval_loss": 0.7240011692047119,
|
| 2435 |
+
"eval_runtime": 6.0767,
|
| 2436 |
+
"eval_samples_per_second": 47.065,
|
| 2437 |
+
"eval_steps_per_second": 2.962,
|
| 2438 |
+
"step": 4753
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 196.99,
|
| 2442 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2443 |
+
"eval_loss": 0.7131750583648682,
|
| 2444 |
+
"eval_runtime": 5.3063,
|
| 2445 |
+
"eval_samples_per_second": 53.898,
|
| 2446 |
+
"eval_steps_per_second": 3.392,
|
| 2447 |
+
"step": 4777
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 197.94,
|
| 2451 |
+
"grad_norm": 0.8933289051055908,
|
| 2452 |
+
"learning_rate": 6.677777777777778e-06,
|
| 2453 |
+
"loss": 0.2116,
|
| 2454 |
+
"step": 4800
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 197.98,
|
| 2458 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2459 |
+
"eval_loss": 0.7169559597969055,
|
| 2460 |
+
"eval_runtime": 5.5713,
|
| 2461 |
+
"eval_samples_per_second": 51.335,
|
| 2462 |
+
"eval_steps_per_second": 3.231,
|
| 2463 |
+
"step": 4801
|
| 2464 |
+
},
|
| 2465 |
+
{
|
| 2466 |
+
"epoch": 198.97,
|
| 2467 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2468 |
+
"eval_loss": 0.7265609502792358,
|
| 2469 |
+
"eval_runtime": 4.1397,
|
| 2470 |
+
"eval_samples_per_second": 69.087,
|
| 2471 |
+
"eval_steps_per_second": 4.348,
|
| 2472 |
+
"step": 4825
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 200.0,
|
| 2476 |
+
"grad_norm": 2.175414562225342,
|
| 2477 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 2478 |
+
"loss": 0.2219,
|
| 2479 |
+
"step": 4850
|
| 2480 |
+
},
|
| 2481 |
+
{
|
| 2482 |
+
"epoch": 200.0,
|
| 2483 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2484 |
+
"eval_loss": 0.7162622213363647,
|
| 2485 |
+
"eval_runtime": 5.2016,
|
| 2486 |
+
"eval_samples_per_second": 54.984,
|
| 2487 |
+
"eval_steps_per_second": 3.461,
|
| 2488 |
+
"step": 4850
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 200.99,
|
| 2492 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2493 |
+
"eval_loss": 0.7302048802375793,
|
| 2494 |
+
"eval_runtime": 4.9222,
|
| 2495 |
+
"eval_samples_per_second": 58.104,
|
| 2496 |
+
"eval_steps_per_second": 3.657,
|
| 2497 |
+
"step": 4874
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 201.98,
|
| 2501 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2502 |
+
"eval_loss": 0.7223746180534363,
|
| 2503 |
+
"eval_runtime": 4.6884,
|
| 2504 |
+
"eval_samples_per_second": 61.002,
|
| 2505 |
+
"eval_steps_per_second": 3.839,
|
| 2506 |
+
"step": 4898
|
| 2507 |
+
},
|
| 2508 |
+
{
|
| 2509 |
+
"epoch": 202.06,
|
| 2510 |
+
"grad_norm": 2.053739309310913,
|
| 2511 |
+
"learning_rate": 6.1222222222222224e-06,
|
| 2512 |
+
"loss": 0.2183,
|
| 2513 |
+
"step": 4900
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 202.97,
|
| 2517 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2518 |
+
"eval_loss": 0.7179226279258728,
|
| 2519 |
+
"eval_runtime": 4.5556,
|
| 2520 |
+
"eval_samples_per_second": 62.78,
|
| 2521 |
+
"eval_steps_per_second": 3.951,
|
| 2522 |
+
"step": 4922
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 204.0,
|
| 2526 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2527 |
+
"eval_loss": 0.7245286107063293,
|
| 2528 |
+
"eval_runtime": 5.7474,
|
| 2529 |
+
"eval_samples_per_second": 49.762,
|
| 2530 |
+
"eval_steps_per_second": 3.132,
|
| 2531 |
+
"step": 4947
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 204.12,
|
| 2535 |
+
"grad_norm": 1.1081063747406006,
|
| 2536 |
+
"learning_rate": 5.844444444444444e-06,
|
| 2537 |
+
"loss": 0.2053,
|
| 2538 |
+
"step": 4950
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 204.99,
|
| 2542 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2543 |
+
"eval_loss": 0.7344977259635925,
|
| 2544 |
+
"eval_runtime": 5.4178,
|
| 2545 |
+
"eval_samples_per_second": 52.789,
|
| 2546 |
+
"eval_steps_per_second": 3.322,
|
| 2547 |
+
"step": 4971
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 205.98,
|
| 2551 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2552 |
+
"eval_loss": 0.7249557971954346,
|
| 2553 |
+
"eval_runtime": 5.6352,
|
| 2554 |
+
"eval_samples_per_second": 50.753,
|
| 2555 |
+
"eval_steps_per_second": 3.194,
|
| 2556 |
+
"step": 4995
|
| 2557 |
+
},
|
| 2558 |
+
{
|
| 2559 |
+
"epoch": 206.19,
|
| 2560 |
+
"grad_norm": 1.09213125705719,
|
| 2561 |
+
"learning_rate": 5.566666666666667e-06,
|
| 2562 |
+
"loss": 0.2113,
|
| 2563 |
+
"step": 5000
|
| 2564 |
+
},
|
| 2565 |
+
{
|
| 2566 |
+
"epoch": 206.97,
|
| 2567 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2568 |
+
"eval_loss": 0.7246001958847046,
|
| 2569 |
+
"eval_runtime": 4.9071,
|
| 2570 |
+
"eval_samples_per_second": 58.283,
|
| 2571 |
+
"eval_steps_per_second": 3.668,
|
| 2572 |
+
"step": 5019
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 208.0,
|
| 2576 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2577 |
+
"eval_loss": 0.7270117998123169,
|
| 2578 |
+
"eval_runtime": 5.8385,
|
| 2579 |
+
"eval_samples_per_second": 48.985,
|
| 2580 |
+
"eval_steps_per_second": 3.083,
|
| 2581 |
+
"step": 5044
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 208.25,
|
| 2585 |
+
"grad_norm": 1.6693130731582642,
|
| 2586 |
+
"learning_rate": 5.288888888888889e-06,
|
| 2587 |
+
"loss": 0.2152,
|
| 2588 |
+
"step": 5050
|
| 2589 |
+
},
|
| 2590 |
+
{
|
| 2591 |
+
"epoch": 208.99,
|
| 2592 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2593 |
+
"eval_loss": 0.7285901308059692,
|
| 2594 |
+
"eval_runtime": 5.489,
|
| 2595 |
+
"eval_samples_per_second": 52.104,
|
| 2596 |
+
"eval_steps_per_second": 3.279,
|
| 2597 |
+
"step": 5068
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 209.98,
|
| 2601 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2602 |
+
"eval_loss": 0.7332947254180908,
|
| 2603 |
+
"eval_runtime": 5.3017,
|
| 2604 |
+
"eval_samples_per_second": 53.945,
|
| 2605 |
+
"eval_steps_per_second": 3.395,
|
| 2606 |
+
"step": 5092
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 210.31,
|
| 2610 |
+
"grad_norm": 2.0511515140533447,
|
| 2611 |
+
"learning_rate": 5.011111111111112e-06,
|
| 2612 |
+
"loss": 0.2129,
|
| 2613 |
+
"step": 5100
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 210.97,
|
| 2617 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2618 |
+
"eval_loss": 0.7307863831520081,
|
| 2619 |
+
"eval_runtime": 5.2991,
|
| 2620 |
+
"eval_samples_per_second": 53.971,
|
| 2621 |
+
"eval_steps_per_second": 3.397,
|
| 2622 |
+
"step": 5116
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 212.0,
|
| 2626 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2627 |
+
"eval_loss": 0.7176437973976135,
|
| 2628 |
+
"eval_runtime": 4.9452,
|
| 2629 |
+
"eval_samples_per_second": 57.834,
|
| 2630 |
+
"eval_steps_per_second": 3.64,
|
| 2631 |
+
"step": 5141
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"epoch": 212.37,
|
| 2635 |
+
"grad_norm": 1.8491023778915405,
|
| 2636 |
+
"learning_rate": 4.7333333333333335e-06,
|
| 2637 |
+
"loss": 0.2173,
|
| 2638 |
+
"step": 5150
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 212.99,
|
| 2642 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2643 |
+
"eval_loss": 0.7334882020950317,
|
| 2644 |
+
"eval_runtime": 4.9602,
|
| 2645 |
+
"eval_samples_per_second": 57.659,
|
| 2646 |
+
"eval_steps_per_second": 3.629,
|
| 2647 |
+
"step": 5165
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"epoch": 213.98,
|
| 2651 |
+
"eval_accuracy": 0.7797202797202797,
|
| 2652 |
+
"eval_loss": 0.7268483638763428,
|
| 2653 |
+
"eval_runtime": 5.885,
|
| 2654 |
+
"eval_samples_per_second": 48.598,
|
| 2655 |
+
"eval_steps_per_second": 3.059,
|
| 2656 |
+
"step": 5189
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 214.43,
|
| 2660 |
+
"grad_norm": 1.2067769765853882,
|
| 2661 |
+
"learning_rate": 4.455555555555556e-06,
|
| 2662 |
+
"loss": 0.2042,
|
| 2663 |
+
"step": 5200
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 214.97,
|
| 2667 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2668 |
+
"eval_loss": 0.7299237847328186,
|
| 2669 |
+
"eval_runtime": 5.7645,
|
| 2670 |
+
"eval_samples_per_second": 49.614,
|
| 2671 |
+
"eval_steps_per_second": 3.123,
|
| 2672 |
+
"step": 5213
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 216.0,
|
| 2676 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2677 |
+
"eval_loss": 0.7360625863075256,
|
| 2678 |
+
"eval_runtime": 4.7143,
|
| 2679 |
+
"eval_samples_per_second": 60.667,
|
| 2680 |
+
"eval_steps_per_second": 3.818,
|
| 2681 |
+
"step": 5238
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 216.49,
|
| 2685 |
+
"grad_norm": 1.3863427639007568,
|
| 2686 |
+
"learning_rate": 4.177777777777777e-06,
|
| 2687 |
+
"loss": 0.2112,
|
| 2688 |
+
"step": 5250
|
| 2689 |
+
},
|
| 2690 |
+
{
|
| 2691 |
+
"epoch": 216.99,
|
| 2692 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2693 |
+
"eval_loss": 0.723866879940033,
|
| 2694 |
+
"eval_runtime": 5.3445,
|
| 2695 |
+
"eval_samples_per_second": 53.513,
|
| 2696 |
+
"eval_steps_per_second": 3.368,
|
| 2697 |
+
"step": 5262
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 217.98,
|
| 2701 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2702 |
+
"eval_loss": 0.7252445220947266,
|
| 2703 |
+
"eval_runtime": 4.6314,
|
| 2704 |
+
"eval_samples_per_second": 61.753,
|
| 2705 |
+
"eval_steps_per_second": 3.887,
|
| 2706 |
+
"step": 5286
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 218.56,
|
| 2710 |
+
"grad_norm": 1.1177924871444702,
|
| 2711 |
+
"learning_rate": 3.9e-06,
|
| 2712 |
+
"loss": 0.2007,
|
| 2713 |
+
"step": 5300
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 218.97,
|
| 2717 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2718 |
+
"eval_loss": 0.719983696937561,
|
| 2719 |
+
"eval_runtime": 4.865,
|
| 2720 |
+
"eval_samples_per_second": 58.787,
|
| 2721 |
+
"eval_steps_per_second": 3.7,
|
| 2722 |
+
"step": 5310
|
| 2723 |
+
},
|
| 2724 |
+
{
|
| 2725 |
+
"epoch": 220.0,
|
| 2726 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2727 |
+
"eval_loss": 0.7195786237716675,
|
| 2728 |
+
"eval_runtime": 5.5422,
|
| 2729 |
+
"eval_samples_per_second": 51.604,
|
| 2730 |
+
"eval_steps_per_second": 3.248,
|
| 2731 |
+
"step": 5335
|
| 2732 |
+
},
|
| 2733 |
+
{
|
| 2734 |
+
"epoch": 220.62,
|
| 2735 |
+
"grad_norm": 1.413304090499878,
|
| 2736 |
+
"learning_rate": 3.6222222222222226e-06,
|
| 2737 |
+
"loss": 0.2163,
|
| 2738 |
+
"step": 5350
|
| 2739 |
+
},
|
| 2740 |
+
{
|
| 2741 |
+
"epoch": 220.99,
|
| 2742 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2743 |
+
"eval_loss": 0.7309580445289612,
|
| 2744 |
+
"eval_runtime": 5.2512,
|
| 2745 |
+
"eval_samples_per_second": 54.463,
|
| 2746 |
+
"eval_steps_per_second": 3.428,
|
| 2747 |
+
"step": 5359
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 221.98,
|
| 2751 |
+
"eval_accuracy": 0.7867132867132867,
|
| 2752 |
+
"eval_loss": 0.7313971519470215,
|
| 2753 |
+
"eval_runtime": 5.1151,
|
| 2754 |
+
"eval_samples_per_second": 55.913,
|
| 2755 |
+
"eval_steps_per_second": 3.519,
|
| 2756 |
+
"step": 5383
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"epoch": 222.68,
|
| 2760 |
+
"grad_norm": 3.0471901893615723,
|
| 2761 |
+
"learning_rate": 3.3444444444444445e-06,
|
| 2762 |
+
"loss": 0.2141,
|
| 2763 |
+
"step": 5400
|
| 2764 |
+
},
|
| 2765 |
+
{
|
| 2766 |
+
"epoch": 222.97,
|
| 2767 |
+
"eval_accuracy": 0.7832167832167832,
|
| 2768 |
+
"eval_loss": 0.727938175201416,
|
| 2769 |
+
"eval_runtime": 4.6405,
|
| 2770 |
+
"eval_samples_per_second": 61.631,
|
| 2771 |
+
"eval_steps_per_second": 3.879,
|
| 2772 |
+
"step": 5407
|
| 2773 |
+
},
|
| 2774 |
+
{
|
| 2775 |
+
"epoch": 224.0,
|
| 2776 |
+
"eval_accuracy": 0.7902097902097902,
|
| 2777 |
+
"eval_loss": 0.725923478603363,
|
| 2778 |
+
"eval_runtime": 5.0906,
|
| 2779 |
+
"eval_samples_per_second": 56.182,
|
| 2780 |
+
"eval_steps_per_second": 3.536,
|
| 2781 |
+
"step": 5432
|
| 2782 |
+
}
|
| 2783 |
+
],
|
| 2784 |
+
"logging_steps": 50,
|
| 2785 |
+
"max_steps": 6000,
|
| 2786 |
+
"num_input_tokens_seen": 0,
|
| 2787 |
+
"num_train_epochs": 250,
|
| 2788 |
+
"save_steps": 500,
|
| 2789 |
+
"total_flos": 3.037085846065152e+18,
|
| 2790 |
+
"train_batch_size": 16,
|
| 2791 |
+
"trial_name": null,
|
| 2792 |
+
"trial_params": null
|
| 2793 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3218c866c7b253f5d3295edcd44b4197864747f68600a16bb0f3d6f506131fb
|
| 3 |
+
size 4984
|