Training in progress, step 3000, checkpoint
Browse files- checkpoint-3000/config.json +91 -0
- checkpoint-3000/generation_config.json +11 -0
- checkpoint-3000/model.safetensors +3 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scaler.pt +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/trainer_state.json +931 -0
- checkpoint-3000/training_args.bin +3 -0
checkpoint-3000/config.json
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"apply_spec_augment": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"SpeechT5ForTextToSpeech"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"conv_bias": false,
|
| 10 |
+
"conv_dim": [
|
| 11 |
+
512,
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512
|
| 18 |
+
],
|
| 19 |
+
"conv_kernel": [
|
| 20 |
+
10,
|
| 21 |
+
3,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
2,
|
| 26 |
+
2
|
| 27 |
+
],
|
| 28 |
+
"conv_stride": [
|
| 29 |
+
5,
|
| 30 |
+
2,
|
| 31 |
+
2,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2
|
| 36 |
+
],
|
| 37 |
+
"decoder_attention_heads": 12,
|
| 38 |
+
"decoder_ffn_dim": 3072,
|
| 39 |
+
"decoder_layerdrop": 0.1,
|
| 40 |
+
"decoder_layers": 6,
|
| 41 |
+
"decoder_start_token_id": 2,
|
| 42 |
+
"dtype": "float32",
|
| 43 |
+
"encoder_attention_heads": 12,
|
| 44 |
+
"encoder_ffn_dim": 3072,
|
| 45 |
+
"encoder_layerdrop": 0.1,
|
| 46 |
+
"encoder_layers": 12,
|
| 47 |
+
"encoder_max_relative_position": 160,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_norm": "group",
|
| 51 |
+
"feat_proj_dropout": 0.0,
|
| 52 |
+
"guided_attention_loss_num_heads": 2,
|
| 53 |
+
"guided_attention_loss_scale": 10.0,
|
| 54 |
+
"guided_attention_loss_sigma": 0.4,
|
| 55 |
+
"hidden_act": "gelu",
|
| 56 |
+
"hidden_dropout": 0.1,
|
| 57 |
+
"hidden_size": 768,
|
| 58 |
+
"initializer_range": 0.02,
|
| 59 |
+
"is_encoder_decoder": true,
|
| 60 |
+
"layer_norm_eps": 1e-05,
|
| 61 |
+
"mask_feature_length": 10,
|
| 62 |
+
"mask_feature_min_masks": 0,
|
| 63 |
+
"mask_feature_prob": 0.0,
|
| 64 |
+
"mask_time_length": 10,
|
| 65 |
+
"mask_time_min_masks": 2,
|
| 66 |
+
"mask_time_prob": 0.05,
|
| 67 |
+
"max_speech_positions": 1876,
|
| 68 |
+
"max_text_positions": 600,
|
| 69 |
+
"model_type": "speecht5",
|
| 70 |
+
"num_conv_pos_embedding_groups": 16,
|
| 71 |
+
"num_conv_pos_embeddings": 128,
|
| 72 |
+
"num_feat_extract_layers": 7,
|
| 73 |
+
"num_mel_bins": 80,
|
| 74 |
+
"pad_token_id": 1,
|
| 75 |
+
"positional_dropout": 0.1,
|
| 76 |
+
"reduction_factor": 2,
|
| 77 |
+
"scale_embedding": false,
|
| 78 |
+
"speaker_embedding_dim": 512,
|
| 79 |
+
"speech_decoder_postnet_dropout": 0.5,
|
| 80 |
+
"speech_decoder_postnet_kernel": 5,
|
| 81 |
+
"speech_decoder_postnet_layers": 5,
|
| 82 |
+
"speech_decoder_postnet_units": 256,
|
| 83 |
+
"speech_decoder_prenet_dropout": 0.5,
|
| 84 |
+
"speech_decoder_prenet_layers": 2,
|
| 85 |
+
"speech_decoder_prenet_units": 256,
|
| 86 |
+
"tie_word_embeddings": true,
|
| 87 |
+
"transformers_version": "5.19.0.dev0",
|
| 88 |
+
"use_cache": false,
|
| 89 |
+
"use_guided_attention_loss": false,
|
| 90 |
+
"vocab_size": 81
|
| 91 |
+
}
|
checkpoint-3000/generation_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"output_attentions": false,
|
| 7 |
+
"output_hidden_states": false,
|
| 8 |
+
"pad_token_id": 1,
|
| 9 |
+
"transformers_version": "5.19.0.dev0",
|
| 10 |
+
"use_cache": true
|
| 11 |
+
}
|
checkpoint-3000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b493c3a35ff1a1630a193fcef2e097fd7da811f2543c07161ca32d457604e5f
|
| 3 |
+
size 577789320
|
checkpoint-3000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:00f68cf8314499fb1d9ae400a6e8848d732bd3f124ff7f7ce8f6e73362548294
|
| 3 |
+
size 1155772634
|
checkpoint-3000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c51bc3c85dc8c58bc257d6b5da0225abdae0ee9168ce430d8f5e00a1240f0be
|
| 3 |
+
size 14645
|
checkpoint-3000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ac1c46a2776d12775d23d0f587efc112188137ce2140da35bc15d301c9f620e
|
| 3 |
+
size 1383
|
checkpoint-3000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8eeacdcbf60c3231fbee28906346814b82c12f413747b7cb8a9a2acca83c9de
|
| 3 |
+
size 1465
|
checkpoint-3000/trainer_state.json
ADDED
|
@@ -0,0 +1,931 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3000,
|
| 3 |
+
"best_metric": 0.9244788289070129,
|
| 4 |
+
"best_model_checkpoint": "/kaggle/working/tts_processed/checkpoint-3000",
|
| 5 |
+
"epoch": 14.852012383900929,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.1238390092879257,
|
| 14 |
+
"grad_norm": 829.4996337890625,
|
| 15 |
+
"learning_rate": 4.8e-08,
|
| 16 |
+
"loss": 21.73699951171875,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.2476780185758514,
|
| 21 |
+
"grad_norm": 701.2330932617188,
|
| 22 |
+
"learning_rate": 9.8e-08,
|
| 23 |
+
"loss": 21.00220947265625,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.3715170278637771,
|
| 28 |
+
"grad_norm": 631.0809326171875,
|
| 29 |
+
"learning_rate": 1.4799999999999998e-07,
|
| 30 |
+
"loss": 19.772623291015623,
|
| 31 |
+
"step": 75
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.4953560371517028,
|
| 35 |
+
"grad_norm": 556.35791015625,
|
| 36 |
+
"learning_rate": 1.98e-07,
|
| 37 |
+
"loss": 18.46118896484375,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6191950464396285,
|
| 42 |
+
"grad_norm": 536.2819213867188,
|
| 43 |
+
"learning_rate": 2.48e-07,
|
| 44 |
+
"loss": 17.51901611328125,
|
| 45 |
+
"step": 125
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.7430340557275542,
|
| 49 |
+
"grad_norm": 459.290283203125,
|
| 50 |
+
"learning_rate": 2.98e-07,
|
| 51 |
+
"loss": 16.43676513671875,
|
| 52 |
+
"step": 150
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.8668730650154799,
|
| 56 |
+
"grad_norm": 230.56405639648438,
|
| 57 |
+
"learning_rate": 3.4799999999999994e-07,
|
| 58 |
+
"loss": 15.75008056640625,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.9907120743034056,
|
| 63 |
+
"grad_norm": 233.90635681152344,
|
| 64 |
+
"learning_rate": 3.98e-07,
|
| 65 |
+
"loss": 14.56845947265625,
|
| 66 |
+
"step": 200
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.1139318885448917,
|
| 70 |
+
"grad_norm": 107.61080169677734,
|
| 71 |
+
"learning_rate": 4.48e-07,
|
| 72 |
+
"loss": 13.99685546875,
|
| 73 |
+
"step": 225
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.2377708978328172,
|
| 77 |
+
"grad_norm": 115.48291015625,
|
| 78 |
+
"learning_rate": 4.979999999999999e-07,
|
| 79 |
+
"loss": 13.390751953125,
|
| 80 |
+
"step": 250
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.361609907120743,
|
| 84 |
+
"grad_norm": 150.88845825195312,
|
| 85 |
+
"learning_rate": 5.48e-07,
|
| 86 |
+
"loss": 12.885958251953125,
|
| 87 |
+
"step": 275
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.4854489164086688,
|
| 91 |
+
"grad_norm": 102.25688934326172,
|
| 92 |
+
"learning_rate": 5.979999999999999e-07,
|
| 93 |
+
"loss": 12.281358642578125,
|
| 94 |
+
"step": 300
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.6092879256965944,
|
| 98 |
+
"grad_norm": 91.96704864501953,
|
| 99 |
+
"learning_rate": 6.48e-07,
|
| 100 |
+
"loss": 12.51485107421875,
|
| 101 |
+
"step": 325
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.73312693498452,
|
| 105 |
+
"grad_norm": 89.19316864013672,
|
| 106 |
+
"learning_rate": 6.979999999999999e-07,
|
| 107 |
+
"loss": 11.758416748046875,
|
| 108 |
+
"step": 350
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.8569659442724458,
|
| 112 |
+
"grad_norm": 43.06995391845703,
|
| 113 |
+
"learning_rate": 7.48e-07,
|
| 114 |
+
"loss": 11.7668603515625,
|
| 115 |
+
"step": 375
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.9808049535603716,
|
| 119 |
+
"grad_norm": 37.787078857421875,
|
| 120 |
+
"learning_rate": 7.98e-07,
|
| 121 |
+
"loss": 11.696016845703125,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.1040247678018575,
|
| 126 |
+
"grad_norm": 52.63007354736328,
|
| 127 |
+
"learning_rate": 8.48e-07,
|
| 128 |
+
"loss": 11.60255859375,
|
| 129 |
+
"step": 425
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.2278637770897833,
|
| 133 |
+
"grad_norm": 37.73246383666992,
|
| 134 |
+
"learning_rate": 8.98e-07,
|
| 135 |
+
"loss": 11.121099853515625,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.351702786377709,
|
| 140 |
+
"grad_norm": 50.654396057128906,
|
| 141 |
+
"learning_rate": 9.479999999999999e-07,
|
| 142 |
+
"loss": 10.981490478515624,
|
| 143 |
+
"step": 475
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.4755417956656345,
|
| 147 |
+
"grad_norm": 34.40877151489258,
|
| 148 |
+
"learning_rate": 9.98e-07,
|
| 149 |
+
"loss": 10.89916015625,
|
| 150 |
+
"step": 500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 2.4755417956656345,
|
| 154 |
+
"eval_loss": 1.2403377294540405,
|
| 155 |
+
"eval_runtime": 78.2214,
|
| 156 |
+
"eval_samples_per_second": 20.659,
|
| 157 |
+
"eval_steps_per_second": 5.165,
|
| 158 |
+
"step": 500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.5993808049535603,
|
| 162 |
+
"grad_norm": 32.53867721557617,
|
| 163 |
+
"learning_rate": 9.986285714285714e-07,
|
| 164 |
+
"loss": 10.70981201171875,
|
| 165 |
+
"step": 525
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.723219814241486,
|
| 169 |
+
"grad_norm": 37.28243637084961,
|
| 170 |
+
"learning_rate": 9.972e-07,
|
| 171 |
+
"loss": 10.68586669921875,
|
| 172 |
+
"step": 550
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.847058823529412,
|
| 176 |
+
"grad_norm": 33.349178314208984,
|
| 177 |
+
"learning_rate": 9.957714285714285e-07,
|
| 178 |
+
"loss": 10.5280859375,
|
| 179 |
+
"step": 575
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.9708978328173377,
|
| 183 |
+
"grad_norm": 40.658817291259766,
|
| 184 |
+
"learning_rate": 9.94342857142857e-07,
|
| 185 |
+
"loss": 10.160640258789062,
|
| 186 |
+
"step": 600
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.0941176470588236,
|
| 190 |
+
"grad_norm": 35.320701599121094,
|
| 191 |
+
"learning_rate": 9.929142857142856e-07,
|
| 192 |
+
"loss": 10.215652465820312,
|
| 193 |
+
"step": 625
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 3.2179566563467494,
|
| 197 |
+
"grad_norm": 31.319137573242188,
|
| 198 |
+
"learning_rate": 9.914857142857142e-07,
|
| 199 |
+
"loss": 10.158941650390625,
|
| 200 |
+
"step": 650
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.341795665634675,
|
| 204 |
+
"grad_norm": 28.438013076782227,
|
| 205 |
+
"learning_rate": 9.900571428571427e-07,
|
| 206 |
+
"loss": 9.933291625976562,
|
| 207 |
+
"step": 675
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.4656346749226006,
|
| 211 |
+
"grad_norm": 41.65836715698242,
|
| 212 |
+
"learning_rate": 9.886285714285713e-07,
|
| 213 |
+
"loss": 9.679817504882813,
|
| 214 |
+
"step": 700
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.5894736842105264,
|
| 218 |
+
"grad_norm": 49.5134162902832,
|
| 219 |
+
"learning_rate": 9.871999999999998e-07,
|
| 220 |
+
"loss": 9.66293212890625,
|
| 221 |
+
"step": 725
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.713312693498452,
|
| 225 |
+
"grad_norm": 37.431331634521484,
|
| 226 |
+
"learning_rate": 9.857714285714284e-07,
|
| 227 |
+
"loss": 9.642633056640625,
|
| 228 |
+
"step": 750
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.837151702786378,
|
| 232 |
+
"grad_norm": 39.974544525146484,
|
| 233 |
+
"learning_rate": 9.843428571428572e-07,
|
| 234 |
+
"loss": 9.296731567382812,
|
| 235 |
+
"step": 775
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 3.9609907120743033,
|
| 239 |
+
"grad_norm": 39.033599853515625,
|
| 240 |
+
"learning_rate": 9.829142857142857e-07,
|
| 241 |
+
"loss": 9.472877807617188,
|
| 242 |
+
"step": 800
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.08421052631579,
|
| 246 |
+
"grad_norm": 29.367778778076172,
|
| 247 |
+
"learning_rate": 9.814857142857143e-07,
|
| 248 |
+
"loss": 9.286629028320313,
|
| 249 |
+
"step": 825
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.208049535603715,
|
| 253 |
+
"grad_norm": 30.283451080322266,
|
| 254 |
+
"learning_rate": 9.800571428571428e-07,
|
| 255 |
+
"loss": 9.21810546875,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 4.33188854489164,
|
| 260 |
+
"grad_norm": 43.98243713378906,
|
| 261 |
+
"learning_rate": 9.786285714285714e-07,
|
| 262 |
+
"loss": 9.4026904296875,
|
| 263 |
+
"step": 875
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 4.455727554179567,
|
| 267 |
+
"grad_norm": 35.94754409790039,
|
| 268 |
+
"learning_rate": 9.772e-07,
|
| 269 |
+
"loss": 9.243385009765625,
|
| 270 |
+
"step": 900
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 4.579566563467492,
|
| 274 |
+
"grad_norm": 97.34254455566406,
|
| 275 |
+
"learning_rate": 9.757714285714285e-07,
|
| 276 |
+
"loss": 9.081926879882813,
|
| 277 |
+
"step": 925
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 4.703405572755418,
|
| 281 |
+
"grad_norm": 35.22529602050781,
|
| 282 |
+
"learning_rate": 9.74342857142857e-07,
|
| 283 |
+
"loss": 9.177518310546875,
|
| 284 |
+
"step": 950
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.827244582043344,
|
| 288 |
+
"grad_norm": 31.761302947998047,
|
| 289 |
+
"learning_rate": 9.729142857142856e-07,
|
| 290 |
+
"loss": 9.266145629882812,
|
| 291 |
+
"step": 975
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.951083591331269,
|
| 295 |
+
"grad_norm": 34.29395294189453,
|
| 296 |
+
"learning_rate": 9.714857142857142e-07,
|
| 297 |
+
"loss": 9.146566162109375,
|
| 298 |
+
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 4.951083591331269,
|
| 302 |
+
"eval_loss": 1.01394784450531,
|
| 303 |
+
"eval_runtime": 78.015,
|
| 304 |
+
"eval_samples_per_second": 20.714,
|
| 305 |
+
"eval_steps_per_second": 5.178,
|
| 306 |
+
"step": 1000
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.074303405572755,
|
| 310 |
+
"grad_norm": 28.847089767456055,
|
| 311 |
+
"learning_rate": 9.700571428571427e-07,
|
| 312 |
+
"loss": 9.07594970703125,
|
| 313 |
+
"step": 1025
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.198142414860681,
|
| 317 |
+
"grad_norm": 31.509410858154297,
|
| 318 |
+
"learning_rate": 9.686285714285713e-07,
|
| 319 |
+
"loss": 9.165030517578124,
|
| 320 |
+
"step": 1050
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.321981424148607,
|
| 324 |
+
"grad_norm": 28.566282272338867,
|
| 325 |
+
"learning_rate": 9.671999999999998e-07,
|
| 326 |
+
"loss": 8.919168701171875,
|
| 327 |
+
"step": 1075
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.445820433436532,
|
| 331 |
+
"grad_norm": 26.087688446044922,
|
| 332 |
+
"learning_rate": 9.657714285714286e-07,
|
| 333 |
+
"loss": 8.877406005859376,
|
| 334 |
+
"step": 1100
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 5.569659442724459,
|
| 338 |
+
"grad_norm": 27.453046798706055,
|
| 339 |
+
"learning_rate": 9.643428571428571e-07,
|
| 340 |
+
"loss": 8.87409912109375,
|
| 341 |
+
"step": 1125
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.693498452012384,
|
| 345 |
+
"grad_norm": 26.87615203857422,
|
| 346 |
+
"learning_rate": 9.629142857142857e-07,
|
| 347 |
+
"loss": 8.98948486328125,
|
| 348 |
+
"step": 1150
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.817337461300309,
|
| 352 |
+
"grad_norm": 28.69729232788086,
|
| 353 |
+
"learning_rate": 9.614857142857143e-07,
|
| 354 |
+
"loss": 8.854622192382813,
|
| 355 |
+
"step": 1175
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.9411764705882355,
|
| 359 |
+
"grad_norm": 31.937828063964844,
|
| 360 |
+
"learning_rate": 9.600571428571428e-07,
|
| 361 |
+
"loss": 9.01401123046875,
|
| 362 |
+
"step": 1200
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 6.064396284829721,
|
| 366 |
+
"grad_norm": 42.471954345703125,
|
| 367 |
+
"learning_rate": 9.586285714285714e-07,
|
| 368 |
+
"loss": 8.731229858398438,
|
| 369 |
+
"step": 1225
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.188235294117647,
|
| 373 |
+
"grad_norm": 23.436538696289062,
|
| 374 |
+
"learning_rate": 9.572e-07,
|
| 375 |
+
"loss": 8.83049560546875,
|
| 376 |
+
"step": 1250
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 6.312074303405573,
|
| 380 |
+
"grad_norm": 36.96808624267578,
|
| 381 |
+
"learning_rate": 9.557714285714285e-07,
|
| 382 |
+
"loss": 8.846947631835938,
|
| 383 |
+
"step": 1275
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 6.435913312693499,
|
| 387 |
+
"grad_norm": 34.232627868652344,
|
| 388 |
+
"learning_rate": 9.54342857142857e-07,
|
| 389 |
+
"loss": 8.8575048828125,
|
| 390 |
+
"step": 1300
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 6.559752321981424,
|
| 394 |
+
"grad_norm": 25.819372177124023,
|
| 395 |
+
"learning_rate": 9.529142857142857e-07,
|
| 396 |
+
"loss": 8.967839965820312,
|
| 397 |
+
"step": 1325
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 6.68359133126935,
|
| 401 |
+
"grad_norm": 30.442367553710938,
|
| 402 |
+
"learning_rate": 9.514857142857142e-07,
|
| 403 |
+
"loss": 8.79695556640625,
|
| 404 |
+
"step": 1350
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.807430340557276,
|
| 408 |
+
"grad_norm": 25.49448585510254,
|
| 409 |
+
"learning_rate": 9.500571428571428e-07,
|
| 410 |
+
"loss": 8.629199829101562,
|
| 411 |
+
"step": 1375
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 6.931269349845201,
|
| 415 |
+
"grad_norm": 50.014862060546875,
|
| 416 |
+
"learning_rate": 9.486285714285714e-07,
|
| 417 |
+
"loss": 8.7688330078125,
|
| 418 |
+
"step": 1400
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 7.054489164086688,
|
| 422 |
+
"grad_norm": 29.534305572509766,
|
| 423 |
+
"learning_rate": 9.472e-07,
|
| 424 |
+
"loss": 8.635714721679687,
|
| 425 |
+
"step": 1425
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 7.178328173374613,
|
| 429 |
+
"grad_norm": 24.548370361328125,
|
| 430 |
+
"learning_rate": 9.457714285714286e-07,
|
| 431 |
+
"loss": 8.751875,
|
| 432 |
+
"step": 1450
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.302167182662538,
|
| 436 |
+
"grad_norm": 27.02566146850586,
|
| 437 |
+
"learning_rate": 9.443428571428571e-07,
|
| 438 |
+
"loss": 8.664575805664063,
|
| 439 |
+
"step": 1475
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 7.4260061919504645,
|
| 443 |
+
"grad_norm": 26.448801040649414,
|
| 444 |
+
"learning_rate": 9.429142857142857e-07,
|
| 445 |
+
"loss": 8.631251831054687,
|
| 446 |
+
"step": 1500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 7.4260061919504645,
|
| 450 |
+
"eval_loss": 0.9785507321357727,
|
| 451 |
+
"eval_runtime": 78.788,
|
| 452 |
+
"eval_samples_per_second": 20.511,
|
| 453 |
+
"eval_steps_per_second": 5.128,
|
| 454 |
+
"step": 1500
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 7.54984520123839,
|
| 458 |
+
"grad_norm": 26.66581916809082,
|
| 459 |
+
"learning_rate": 9.414857142857142e-07,
|
| 460 |
+
"loss": 8.60765869140625,
|
| 461 |
+
"step": 1525
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 7.673684210526316,
|
| 465 |
+
"grad_norm": 24.334070205688477,
|
| 466 |
+
"learning_rate": 9.400571428571428e-07,
|
| 467 |
+
"loss": 8.694551391601562,
|
| 468 |
+
"step": 1550
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 7.7975232198142415,
|
| 472 |
+
"grad_norm": 25.746564865112305,
|
| 473 |
+
"learning_rate": 9.386285714285714e-07,
|
| 474 |
+
"loss": 8.605336303710937,
|
| 475 |
+
"step": 1575
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 7.921362229102167,
|
| 479 |
+
"grad_norm": 26.221471786499023,
|
| 480 |
+
"learning_rate": 9.372e-07,
|
| 481 |
+
"loss": 8.670101928710938,
|
| 482 |
+
"step": 1600
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 8.044582043343654,
|
| 486 |
+
"grad_norm": 40.2961311340332,
|
| 487 |
+
"learning_rate": 9.357714285714286e-07,
|
| 488 |
+
"loss": 8.697719116210937,
|
| 489 |
+
"step": 1625
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 8.16842105263158,
|
| 493 |
+
"grad_norm": 29.41912078857422,
|
| 494 |
+
"learning_rate": 9.343428571428571e-07,
|
| 495 |
+
"loss": 8.6234228515625,
|
| 496 |
+
"step": 1650
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 8.292260061919505,
|
| 500 |
+
"grad_norm": 23.59851837158203,
|
| 501 |
+
"learning_rate": 9.329142857142857e-07,
|
| 502 |
+
"loss": 8.555197143554688,
|
| 503 |
+
"step": 1675
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 8.41609907120743,
|
| 507 |
+
"grad_norm": 23.335365295410156,
|
| 508 |
+
"learning_rate": 9.314857142857142e-07,
|
| 509 |
+
"loss": 8.533560180664063,
|
| 510 |
+
"step": 1700
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 8.539938080495356,
|
| 514 |
+
"grad_norm": 19.63068389892578,
|
| 515 |
+
"learning_rate": 9.300571428571428e-07,
|
| 516 |
+
"loss": 8.534976196289062,
|
| 517 |
+
"step": 1725
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 8.66377708978328,
|
| 521 |
+
"grad_norm": 26.558069229125977,
|
| 522 |
+
"learning_rate": 9.286285714285714e-07,
|
| 523 |
+
"loss": 8.529254150390624,
|
| 524 |
+
"step": 1750
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 8.787616099071208,
|
| 528 |
+
"grad_norm": 28.553796768188477,
|
| 529 |
+
"learning_rate": 9.272e-07,
|
| 530 |
+
"loss": 8.4541943359375,
|
| 531 |
+
"step": 1775
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 8.911455108359133,
|
| 535 |
+
"grad_norm": 26.726491928100586,
|
| 536 |
+
"learning_rate": 9.257714285714285e-07,
|
| 537 |
+
"loss": 8.505891723632812,
|
| 538 |
+
"step": 1800
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 9.034674922600619,
|
| 542 |
+
"grad_norm": 33.665992736816406,
|
| 543 |
+
"learning_rate": 9.243428571428571e-07,
|
| 544 |
+
"loss": 8.420493774414062,
|
| 545 |
+
"step": 1825
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.158513931888544,
|
| 549 |
+
"grad_norm": 26.62388801574707,
|
| 550 |
+
"learning_rate": 9.229142857142857e-07,
|
| 551 |
+
"loss": 8.4884375,
|
| 552 |
+
"step": 1850
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 9.282352941176471,
|
| 556 |
+
"grad_norm": 21.369853973388672,
|
| 557 |
+
"learning_rate": 9.214857142857142e-07,
|
| 558 |
+
"loss": 8.413807983398437,
|
| 559 |
+
"step": 1875
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 9.406191950464397,
|
| 563 |
+
"grad_norm": 23.232192993164062,
|
| 564 |
+
"learning_rate": 9.200571428571429e-07,
|
| 565 |
+
"loss": 8.473324584960938,
|
| 566 |
+
"step": 1900
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.530030959752322,
|
| 570 |
+
"grad_norm": 48.790531158447266,
|
| 571 |
+
"learning_rate": 9.186285714285714e-07,
|
| 572 |
+
"loss": 8.452188720703125,
|
| 573 |
+
"step": 1925
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.653869969040247,
|
| 577 |
+
"grad_norm": 23.798295974731445,
|
| 578 |
+
"learning_rate": 9.172e-07,
|
| 579 |
+
"loss": 8.495001831054687,
|
| 580 |
+
"step": 1950
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 9.777708978328173,
|
| 584 |
+
"grad_norm": 33.404537200927734,
|
| 585 |
+
"learning_rate": 9.157714285714285e-07,
|
| 586 |
+
"loss": 8.58928955078125,
|
| 587 |
+
"step": 1975
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 9.9015479876161,
|
| 591 |
+
"grad_norm": 30.602357864379883,
|
| 592 |
+
"learning_rate": 9.143428571428571e-07,
|
| 593 |
+
"loss": 8.496845703125,
|
| 594 |
+
"step": 2000
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 9.9015479876161,
|
| 598 |
+
"eval_loss": 0.9522278904914856,
|
| 599 |
+
"eval_runtime": 78.3957,
|
| 600 |
+
"eval_samples_per_second": 20.613,
|
| 601 |
+
"eval_steps_per_second": 5.153,
|
| 602 |
+
"step": 2000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 10.024767801857585,
|
| 606 |
+
"grad_norm": 28.26131248474121,
|
| 607 |
+
"learning_rate": 9.129142857142856e-07,
|
| 608 |
+
"loss": 8.348336181640626,
|
| 609 |
+
"step": 2025
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 10.14860681114551,
|
| 613 |
+
"grad_norm": 27.35468864440918,
|
| 614 |
+
"learning_rate": 9.114857142857142e-07,
|
| 615 |
+
"loss": 8.510921020507812,
|
| 616 |
+
"step": 2050
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 10.272445820433436,
|
| 620 |
+
"grad_norm": 21.89419174194336,
|
| 621 |
+
"learning_rate": 9.100571428571429e-07,
|
| 622 |
+
"loss": 8.4816162109375,
|
| 623 |
+
"step": 2075
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 10.396284829721361,
|
| 627 |
+
"grad_norm": 24.475194931030273,
|
| 628 |
+
"learning_rate": 9.086285714285714e-07,
|
| 629 |
+
"loss": 8.524483032226563,
|
| 630 |
+
"step": 2100
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 10.520123839009289,
|
| 634 |
+
"grad_norm": 28.319612503051758,
|
| 635 |
+
"learning_rate": 9.072e-07,
|
| 636 |
+
"loss": 8.434509887695313,
|
| 637 |
+
"step": 2125
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 10.643962848297214,
|
| 641 |
+
"grad_norm": 22.886144638061523,
|
| 642 |
+
"learning_rate": 9.057714285714285e-07,
|
| 643 |
+
"loss": 8.604608764648438,
|
| 644 |
+
"step": 2150
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 10.76780185758514,
|
| 648 |
+
"grad_norm": 31.52608871459961,
|
| 649 |
+
"learning_rate": 9.043428571428571e-07,
|
| 650 |
+
"loss": 8.302666015625,
|
| 651 |
+
"step": 2175
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 10.891640866873065,
|
| 655 |
+
"grad_norm": 23.449596405029297,
|
| 656 |
+
"learning_rate": 9.029142857142856e-07,
|
| 657 |
+
"loss": 8.249719848632813,
|
| 658 |
+
"step": 2200
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 11.014860681114552,
|
| 662 |
+
"grad_norm": 26.614627838134766,
|
| 663 |
+
"learning_rate": 9.014857142857143e-07,
|
| 664 |
+
"loss": 8.312252807617188,
|
| 665 |
+
"step": 2225
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 11.138699690402477,
|
| 669 |
+
"grad_norm": 25.212867736816406,
|
| 670 |
+
"learning_rate": 9.000571428571429e-07,
|
| 671 |
+
"loss": 8.25859130859375,
|
| 672 |
+
"step": 2250
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 11.262538699690403,
|
| 676 |
+
"grad_norm": 27.4830265045166,
|
| 677 |
+
"learning_rate": 8.986285714285714e-07,
|
| 678 |
+
"loss": 8.3023095703125,
|
| 679 |
+
"step": 2275
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 11.386377708978328,
|
| 683 |
+
"grad_norm": 23.062416076660156,
|
| 684 |
+
"learning_rate": 8.972e-07,
|
| 685 |
+
"loss": 8.250548095703126,
|
| 686 |
+
"step": 2300
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 11.510216718266253,
|
| 690 |
+
"grad_norm": 26.304353713989258,
|
| 691 |
+
"learning_rate": 8.957714285714285e-07,
|
| 692 |
+
"loss": 8.221282958984375,
|
| 693 |
+
"step": 2325
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 11.63405572755418,
|
| 697 |
+
"grad_norm": 30.5177001953125,
|
| 698 |
+
"learning_rate": 8.943428571428571e-07,
|
| 699 |
+
"loss": 8.381609497070313,
|
| 700 |
+
"step": 2350
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 11.757894736842106,
|
| 704 |
+
"grad_norm": 23.5266056060791,
|
| 705 |
+
"learning_rate": 8.929142857142856e-07,
|
| 706 |
+
"loss": 8.32532958984375,
|
| 707 |
+
"step": 2375
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 11.881733746130031,
|
| 711 |
+
"grad_norm": 24.466337203979492,
|
| 712 |
+
"learning_rate": 8.914857142857143e-07,
|
| 713 |
+
"loss": 8.319666137695313,
|
| 714 |
+
"step": 2400
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 12.004953560371517,
|
| 718 |
+
"grad_norm": 28.647558212280273,
|
| 719 |
+
"learning_rate": 8.900571428571428e-07,
|
| 720 |
+
"loss": 8.305534057617187,
|
| 721 |
+
"step": 2425
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 12.128792569659442,
|
| 725 |
+
"grad_norm": 25.798559188842773,
|
| 726 |
+
"learning_rate": 8.886285714285714e-07,
|
| 727 |
+
"loss": 8.325819091796875,
|
| 728 |
+
"step": 2450
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.25263157894737,
|
| 732 |
+
"grad_norm": 23.050085067749023,
|
| 733 |
+
"learning_rate": 8.872e-07,
|
| 734 |
+
"loss": 8.418189086914062,
|
| 735 |
+
"step": 2475
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.376470588235295,
|
| 739 |
+
"grad_norm": 45.27325439453125,
|
| 740 |
+
"learning_rate": 8.857714285714285e-07,
|
| 741 |
+
"loss": 8.232376098632812,
|
| 742 |
+
"step": 2500
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 12.376470588235295,
|
| 746 |
+
"eval_loss": 0.9387099742889404,
|
| 747 |
+
"eval_runtime": 78.8561,
|
| 748 |
+
"eval_samples_per_second": 20.493,
|
| 749 |
+
"eval_steps_per_second": 5.123,
|
| 750 |
+
"step": 2500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 12.50030959752322,
|
| 754 |
+
"grad_norm": 30.00120735168457,
|
| 755 |
+
"learning_rate": 8.843428571428571e-07,
|
| 756 |
+
"loss": 8.216619262695312,
|
| 757 |
+
"step": 2525
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 12.624148606811145,
|
| 761 |
+
"grad_norm": 20.330636978149414,
|
| 762 |
+
"learning_rate": 8.829142857142857e-07,
|
| 763 |
+
"loss": 8.305621948242187,
|
| 764 |
+
"step": 2550
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 12.74798761609907,
|
| 768 |
+
"grad_norm": 23.445192337036133,
|
| 769 |
+
"learning_rate": 8.814857142857143e-07,
|
| 770 |
+
"loss": 8.283675537109374,
|
| 771 |
+
"step": 2575
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 12.871826625386998,
|
| 775 |
+
"grad_norm": 19.532381057739258,
|
| 776 |
+
"learning_rate": 8.800571428571428e-07,
|
| 777 |
+
"loss": 8.197587890625,
|
| 778 |
+
"step": 2600
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 12.995665634674923,
|
| 782 |
+
"grad_norm": 24.901432037353516,
|
| 783 |
+
"learning_rate": 8.786285714285714e-07,
|
| 784 |
+
"loss": 8.2815869140625,
|
| 785 |
+
"step": 2625
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 13.118885448916409,
|
| 789 |
+
"grad_norm": 23.698482513427734,
|
| 790 |
+
"learning_rate": 8.771999999999999e-07,
|
| 791 |
+
"loss": 8.184017333984375,
|
| 792 |
+
"step": 2650
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 13.242724458204334,
|
| 796 |
+
"grad_norm": 22.537649154663086,
|
| 797 |
+
"learning_rate": 8.757714285714285e-07,
|
| 798 |
+
"loss": 8.187122802734375,
|
| 799 |
+
"step": 2675
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 13.36656346749226,
|
| 803 |
+
"grad_norm": 23.588315963745117,
|
| 804 |
+
"learning_rate": 8.74342857142857e-07,
|
| 805 |
+
"loss": 8.233043212890625,
|
| 806 |
+
"step": 2700
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 13.490402476780186,
|
| 810 |
+
"grad_norm": 31.038589477539062,
|
| 811 |
+
"learning_rate": 8.729142857142857e-07,
|
| 812 |
+
"loss": 8.288897094726563,
|
| 813 |
+
"step": 2725
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 13.614241486068112,
|
| 817 |
+
"grad_norm": 27.92134666442871,
|
| 818 |
+
"learning_rate": 8.714857142857143e-07,
|
| 819 |
+
"loss": 8.241436767578126,
|
| 820 |
+
"step": 2750
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 13.738080495356037,
|
| 824 |
+
"grad_norm": 22.192489624023438,
|
| 825 |
+
"learning_rate": 8.700571428571428e-07,
|
| 826 |
+
"loss": 8.224349975585938,
|
| 827 |
+
"step": 2775
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.861919504643962,
|
| 831 |
+
"grad_norm": 29.820369720458984,
|
| 832 |
+
"learning_rate": 8.686285714285714e-07,
|
| 833 |
+
"loss": 8.162620849609375,
|
| 834 |
+
"step": 2800
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 13.985758513931888,
|
| 838 |
+
"grad_norm": 28.781845092773438,
|
| 839 |
+
"learning_rate": 8.671999999999999e-07,
|
| 840 |
+
"loss": 8.23837890625,
|
| 841 |
+
"step": 2825
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 14.108978328173375,
|
| 845 |
+
"grad_norm": 21.96494483947754,
|
| 846 |
+
"learning_rate": 8.657714285714285e-07,
|
| 847 |
+
"loss": 8.158751831054687,
|
| 848 |
+
"step": 2850
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 14.2328173374613,
|
| 852 |
+
"grad_norm": 22.800256729125977,
|
| 853 |
+
"learning_rate": 8.643428571428571e-07,
|
| 854 |
+
"loss": 8.228921508789062,
|
| 855 |
+
"step": 2875
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 14.356656346749226,
|
| 859 |
+
"grad_norm": 23.881921768188477,
|
| 860 |
+
"learning_rate": 8.629142857142857e-07,
|
| 861 |
+
"loss": 8.183645629882813,
|
| 862 |
+
"step": 2900
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 14.480495356037151,
|
| 866 |
+
"grad_norm": 22.399667739868164,
|
| 867 |
+
"learning_rate": 8.614857142857143e-07,
|
| 868 |
+
"loss": 8.157160034179688,
|
| 869 |
+
"step": 2925
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 14.604334365325077,
|
| 873 |
+
"grad_norm": 22.374099731445312,
|
| 874 |
+
"learning_rate": 8.600571428571428e-07,
|
| 875 |
+
"loss": 8.129403076171876,
|
| 876 |
+
"step": 2950
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 14.728173374613004,
|
| 880 |
+
"grad_norm": 22.495939254760742,
|
| 881 |
+
"learning_rate": 8.586285714285714e-07,
|
| 882 |
+
"loss": 8.19878662109375,
|
| 883 |
+
"step": 2975
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 14.852012383900929,
|
| 887 |
+
"grad_norm": 26.343931198120117,
|
| 888 |
+
"learning_rate": 8.571999999999999e-07,
|
| 889 |
+
"loss": 8.154603881835937,
|
| 890 |
+
"step": 3000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 14.852012383900929,
|
| 894 |
+
"eval_loss": 0.9244788289070129,
|
| 895 |
+
"eval_runtime": 78.6746,
|
| 896 |
+
"eval_samples_per_second": 20.54,
|
| 897 |
+
"eval_steps_per_second": 5.135,
|
| 898 |
+
"step": 3000
|
| 899 |
+
}
|
| 900 |
+
],
|
| 901 |
+
"logging_steps": 25,
|
| 902 |
+
"max_steps": 18000,
|
| 903 |
+
"num_input_tokens_seen": 0,
|
| 904 |
+
"num_train_epochs": 90,
|
| 905 |
+
"save_steps": 500,
|
| 906 |
+
"stateful_callbacks": {
|
| 907 |
+
"EarlyStoppingCallback": {
|
| 908 |
+
"args": {
|
| 909 |
+
"early_stopping_patience": 5,
|
| 910 |
+
"early_stopping_threshold": 0.0
|
| 911 |
+
},
|
| 912 |
+
"attributes": {
|
| 913 |
+
"early_stopping_patience_counter": 0
|
| 914 |
+
}
|
| 915 |
+
},
|
| 916 |
+
"TrainerControl": {
|
| 917 |
+
"args": {
|
| 918 |
+
"should_epoch_stop": false,
|
| 919 |
+
"should_evaluate": false,
|
| 920 |
+
"should_log": false,
|
| 921 |
+
"should_save": true,
|
| 922 |
+
"should_training_stop": false
|
| 923 |
+
},
|
| 924 |
+
"attributes": {}
|
| 925 |
+
}
|
| 926 |
+
},
|
| 927 |
+
"total_flos": 1.1655819977114016e+16,
|
| 928 |
+
"train_batch_size": 4,
|
| 929 |
+
"trial_name": null,
|
| 930 |
+
"trial_params": null
|
| 931 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc3ae1f3188fc78f14b1e24c35f0f902ff13c8475d4503f269d542bb33a09e53
|
| 3 |
+
size 5393
|