Training in progress, epoch 1, checkpoint
Browse files- checkpoint-490/config.json +71 -0
- checkpoint-490/model.safetensors +3 -0
- checkpoint-490/optimizer.pt +3 -0
- checkpoint-490/preprocessor_config.json +14 -0
- checkpoint-490/rng_state.pth +3 -0
- checkpoint-490/scheduler.pt +3 -0
- checkpoint-490/trainer_state.json +728 -0
- checkpoint-490/training_args.bin +3 -0
checkpoint-490/config.json
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vinai/PhoWhisper-small",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"apply_spec_augment": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"ViSpeechClassification"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"begin_suppress_tokens": [
|
| 11 |
+
220,
|
| 12 |
+
50257
|
| 13 |
+
],
|
| 14 |
+
"bos_token_id": 50257,
|
| 15 |
+
"classifier_proj_size": 256,
|
| 16 |
+
"d_model": 768,
|
| 17 |
+
"decoder_attention_heads": 12,
|
| 18 |
+
"decoder_ffn_dim": 3072,
|
| 19 |
+
"decoder_layerdrop": 0.0,
|
| 20 |
+
"decoder_layers": 12,
|
| 21 |
+
"decoder_start_token_id": 50258,
|
| 22 |
+
"dropout": 0.0,
|
| 23 |
+
"encoder_attention_heads": 12,
|
| 24 |
+
"encoder_ffn_dim": 3072,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 50257,
|
| 28 |
+
"forced_decoder_ids": null,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "male, northern dialect",
|
| 31 |
+
"1": "male, central dialect",
|
| 32 |
+
"2": "male, highland central dialect",
|
| 33 |
+
"3": "male, southern dialect",
|
| 34 |
+
"4": "female, northern dialect",
|
| 35 |
+
"5": "female, central dialect",
|
| 36 |
+
"6": "female, highland central dialect",
|
| 37 |
+
"7": "female, southern dialect"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"female, central dialect": "5",
|
| 43 |
+
"female, highland central dialect": "6",
|
| 44 |
+
"female, northern dialect": "4",
|
| 45 |
+
"female, southern dialect": "7",
|
| 46 |
+
"male, central dialect": "1",
|
| 47 |
+
"male, highland central dialect": "2",
|
| 48 |
+
"male, northern dialect": "0",
|
| 49 |
+
"male, southern dialect": "3"
|
| 50 |
+
},
|
| 51 |
+
"mask_feature_length": 10,
|
| 52 |
+
"mask_feature_min_masks": 0,
|
| 53 |
+
"mask_feature_prob": 0.0,
|
| 54 |
+
"mask_time_length": 10,
|
| 55 |
+
"mask_time_min_masks": 2,
|
| 56 |
+
"mask_time_prob": 0.05,
|
| 57 |
+
"max_length": 448,
|
| 58 |
+
"max_source_positions": 1500,
|
| 59 |
+
"max_target_positions": 448,
|
| 60 |
+
"median_filter_width": 7,
|
| 61 |
+
"model_type": "whisper",
|
| 62 |
+
"num_hidden_layers": 12,
|
| 63 |
+
"num_mel_bins": 80,
|
| 64 |
+
"pad_token_id": 50257,
|
| 65 |
+
"scale_embedding": false,
|
| 66 |
+
"torch_dtype": "float32",
|
| 67 |
+
"transformers_version": "4.41.2",
|
| 68 |
+
"use_cache": true,
|
| 69 |
+
"use_weighted_layer_sum": false,
|
| 70 |
+
"vocab_size": 51865
|
| 71 |
+
}
|
checkpoint-490/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1f83a195b17329df965f502e1dbc02ba8e219d67b4af419bee6e3cc9ccd4a81
|
| 3 |
+
size 358419776
|
checkpoint-490/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a193a5c3ac38afe35ad02511643dcf4c87975a91803b8438a4124c28c28d6edb
|
| 3 |
+
size 707744314
|
checkpoint-490/preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 4 |
+
"feature_size": 80,
|
| 5 |
+
"hop_length": 160,
|
| 6 |
+
"n_fft": 400,
|
| 7 |
+
"n_samples": 480000,
|
| 8 |
+
"nb_max_frames": 3000,
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"padding_value": 0.0,
|
| 11 |
+
"processor_class": "WhisperProcessor",
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
checkpoint-490/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa339e69c89f04f664bc5b5af50ad45cd97bd253a6b751985062b9e53256a30b
|
| 3 |
+
size 14244
|
checkpoint-490/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:245145eb9a6fd489c0d69d45ffebd818bbeab6ceb70ee100baa6caa724160cad
|
| 3 |
+
size 1064
|
checkpoint-490/trainer_state.json
ADDED
|
@@ -0,0 +1,728 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4966465459423206,
|
| 3 |
+
"best_model_checkpoint": "PhoWhisper-small-vispeech-classifier-v3/checkpoint-490",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 490,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01020408163265306,
|
| 13 |
+
"grad_norm": 58959.5,
|
| 14 |
+
"learning_rate": 1.7006802721088437e-08,
|
| 15 |
+
"loss": 2.0854,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02040816326530612,
|
| 20 |
+
"grad_norm": 54177.4375,
|
| 21 |
+
"learning_rate": 3.4013605442176873e-08,
|
| 22 |
+
"loss": 2.0829,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.030612244897959183,
|
| 27 |
+
"grad_norm": 47591.8828125,
|
| 28 |
+
"learning_rate": 5.1020408163265303e-08,
|
| 29 |
+
"loss": 2.0833,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04081632653061224,
|
| 34 |
+
"grad_norm": 54382.19921875,
|
| 35 |
+
"learning_rate": 6.802721088435375e-08,
|
| 36 |
+
"loss": 2.0864,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05102040816326531,
|
| 41 |
+
"grad_norm": 66346.9609375,
|
| 42 |
+
"learning_rate": 8.50340136054422e-08,
|
| 43 |
+
"loss": 2.0852,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.061224489795918366,
|
| 48 |
+
"grad_norm": 57170.58203125,
|
| 49 |
+
"learning_rate": 1.0204081632653061e-07,
|
| 50 |
+
"loss": 2.0855,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07142857142857142,
|
| 55 |
+
"grad_norm": 54366.90625,
|
| 56 |
+
"learning_rate": 1.1904761904761906e-07,
|
| 57 |
+
"loss": 2.0841,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08163265306122448,
|
| 62 |
+
"grad_norm": 62029.41796875,
|
| 63 |
+
"learning_rate": 1.360544217687075e-07,
|
| 64 |
+
"loss": 2.082,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09183673469387756,
|
| 69 |
+
"grad_norm": 58306.72265625,
|
| 70 |
+
"learning_rate": 1.5306122448979592e-07,
|
| 71 |
+
"loss": 2.0836,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10204081632653061,
|
| 76 |
+
"grad_norm": 53709.69921875,
|
| 77 |
+
"learning_rate": 1.700680272108844e-07,
|
| 78 |
+
"loss": 2.0841,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11224489795918367,
|
| 83 |
+
"grad_norm": 74088.328125,
|
| 84 |
+
"learning_rate": 1.8707482993197282e-07,
|
| 85 |
+
"loss": 2.0825,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12244897959183673,
|
| 90 |
+
"grad_norm": 57089.6796875,
|
| 91 |
+
"learning_rate": 2.0408163265306121e-07,
|
| 92 |
+
"loss": 2.0799,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1326530612244898,
|
| 97 |
+
"grad_norm": 60232.51953125,
|
| 98 |
+
"learning_rate": 2.210884353741497e-07,
|
| 99 |
+
"loss": 2.0791,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14285714285714285,
|
| 104 |
+
"grad_norm": 56240.4609375,
|
| 105 |
+
"learning_rate": 2.3809523809523811e-07,
|
| 106 |
+
"loss": 2.0828,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15306122448979592,
|
| 111 |
+
"grad_norm": 74859.0859375,
|
| 112 |
+
"learning_rate": 2.5510204081632656e-07,
|
| 113 |
+
"loss": 2.0817,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.16326530612244897,
|
| 118 |
+
"grad_norm": 57771.18359375,
|
| 119 |
+
"learning_rate": 2.72108843537415e-07,
|
| 120 |
+
"loss": 2.0796,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17346938775510204,
|
| 125 |
+
"grad_norm": 64205.13671875,
|
| 126 |
+
"learning_rate": 2.891156462585034e-07,
|
| 127 |
+
"loss": 2.0785,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.1836734693877551,
|
| 132 |
+
"grad_norm": 70020.09375,
|
| 133 |
+
"learning_rate": 3.0612244897959183e-07,
|
| 134 |
+
"loss": 2.0784,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.19387755102040816,
|
| 139 |
+
"grad_norm": 55830.03125,
|
| 140 |
+
"learning_rate": 3.231292517006803e-07,
|
| 141 |
+
"loss": 2.0786,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.20408163265306123,
|
| 146 |
+
"grad_norm": 62736.1328125,
|
| 147 |
+
"learning_rate": 3.401360544217688e-07,
|
| 148 |
+
"loss": 2.0761,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21428571428571427,
|
| 153 |
+
"grad_norm": 48753.46875,
|
| 154 |
+
"learning_rate": 3.5714285714285716e-07,
|
| 155 |
+
"loss": 2.0777,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22448979591836735,
|
| 160 |
+
"grad_norm": 70695.0078125,
|
| 161 |
+
"learning_rate": 3.7414965986394563e-07,
|
| 162 |
+
"loss": 2.0776,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.23469387755102042,
|
| 167 |
+
"grad_norm": 55496.5078125,
|
| 168 |
+
"learning_rate": 3.9115646258503406e-07,
|
| 169 |
+
"loss": 2.0755,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.24489795918367346,
|
| 174 |
+
"grad_norm": 82219.4375,
|
| 175 |
+
"learning_rate": 4.0816326530612243e-07,
|
| 176 |
+
"loss": 2.0698,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25510204081632654,
|
| 181 |
+
"grad_norm": 69695.9375,
|
| 182 |
+
"learning_rate": 4.251700680272109e-07,
|
| 183 |
+
"loss": 2.0722,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2653061224489796,
|
| 188 |
+
"grad_norm": 73932.9453125,
|
| 189 |
+
"learning_rate": 4.421768707482994e-07,
|
| 190 |
+
"loss": 2.0718,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2755102040816326,
|
| 195 |
+
"grad_norm": 67854.5859375,
|
| 196 |
+
"learning_rate": 4.591836734693878e-07,
|
| 197 |
+
"loss": 2.07,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2857142857142857,
|
| 202 |
+
"grad_norm": 53210.015625,
|
| 203 |
+
"learning_rate": 4.7619047619047623e-07,
|
| 204 |
+
"loss": 2.0712,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.29591836734693877,
|
| 209 |
+
"grad_norm": 62421.94140625,
|
| 210 |
+
"learning_rate": 4.931972789115647e-07,
|
| 211 |
+
"loss": 2.0736,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.30612244897959184,
|
| 216 |
+
"grad_norm": 77439.4453125,
|
| 217 |
+
"learning_rate": 5.102040816326531e-07,
|
| 218 |
+
"loss": 2.0666,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3163265306122449,
|
| 223 |
+
"grad_norm": 56657.57421875,
|
| 224 |
+
"learning_rate": 5.272108843537416e-07,
|
| 225 |
+
"loss": 2.0646,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.32653061224489793,
|
| 230 |
+
"grad_norm": 74708.640625,
|
| 231 |
+
"learning_rate": 5.4421768707483e-07,
|
| 232 |
+
"loss": 2.0625,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.336734693877551,
|
| 237 |
+
"grad_norm": 48720.30078125,
|
| 238 |
+
"learning_rate": 5.612244897959184e-07,
|
| 239 |
+
"loss": 2.0658,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3469387755102041,
|
| 244 |
+
"grad_norm": 51835.54296875,
|
| 245 |
+
"learning_rate": 5.782312925170068e-07,
|
| 246 |
+
"loss": 2.0615,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35714285714285715,
|
| 251 |
+
"grad_norm": 45173.23046875,
|
| 252 |
+
"learning_rate": 5.952380952380953e-07,
|
| 253 |
+
"loss": 2.0602,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.3673469387755102,
|
| 258 |
+
"grad_norm": 57442.7265625,
|
| 259 |
+
"learning_rate": 6.122448979591837e-07,
|
| 260 |
+
"loss": 2.0592,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.37755102040816324,
|
| 265 |
+
"grad_norm": 77342.5,
|
| 266 |
+
"learning_rate": 6.292517006802721e-07,
|
| 267 |
+
"loss": 2.0581,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.3877551020408163,
|
| 272 |
+
"grad_norm": 75636.84375,
|
| 273 |
+
"learning_rate": 6.462585034013606e-07,
|
| 274 |
+
"loss": 2.0556,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3979591836734694,
|
| 279 |
+
"grad_norm": 58702.5703125,
|
| 280 |
+
"learning_rate": 6.632653061224491e-07,
|
| 281 |
+
"loss": 2.0579,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.40816326530612246,
|
| 286 |
+
"grad_norm": 66054.546875,
|
| 287 |
+
"learning_rate": 6.802721088435376e-07,
|
| 288 |
+
"loss": 2.0559,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.41836734693877553,
|
| 293 |
+
"grad_norm": 62451.17578125,
|
| 294 |
+
"learning_rate": 6.972789115646258e-07,
|
| 295 |
+
"loss": 2.0471,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.42857142857142855,
|
| 300 |
+
"grad_norm": 53168.203125,
|
| 301 |
+
"learning_rate": 7.142857142857143e-07,
|
| 302 |
+
"loss": 2.0459,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.4387755102040816,
|
| 307 |
+
"grad_norm": 67945.484375,
|
| 308 |
+
"learning_rate": 7.312925170068028e-07,
|
| 309 |
+
"loss": 2.0444,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4489795918367347,
|
| 314 |
+
"grad_norm": 51722.9140625,
|
| 315 |
+
"learning_rate": 7.482993197278913e-07,
|
| 316 |
+
"loss": 2.0402,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.45918367346938777,
|
| 321 |
+
"grad_norm": 74699.6796875,
|
| 322 |
+
"learning_rate": 7.653061224489796e-07,
|
| 323 |
+
"loss": 2.0316,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.46938775510204084,
|
| 328 |
+
"grad_norm": 41176.36328125,
|
| 329 |
+
"learning_rate": 7.823129251700681e-07,
|
| 330 |
+
"loss": 2.0442,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.47959183673469385,
|
| 335 |
+
"grad_norm": 56116.875,
|
| 336 |
+
"learning_rate": 7.993197278911566e-07,
|
| 337 |
+
"loss": 2.0403,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4897959183673469,
|
| 342 |
+
"grad_norm": 69240.3359375,
|
| 343 |
+
"learning_rate": 8.163265306122449e-07,
|
| 344 |
+
"loss": 2.037,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5,
|
| 349 |
+
"grad_norm": 67816.90625,
|
| 350 |
+
"learning_rate": 8.333333333333333e-07,
|
| 351 |
+
"loss": 2.0273,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5102040816326531,
|
| 356 |
+
"grad_norm": 64964.8515625,
|
| 357 |
+
"learning_rate": 8.503401360544218e-07,
|
| 358 |
+
"loss": 2.0358,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5204081632653061,
|
| 363 |
+
"grad_norm": 75715.4765625,
|
| 364 |
+
"learning_rate": 8.673469387755103e-07,
|
| 365 |
+
"loss": 2.0108,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5306122448979592,
|
| 370 |
+
"grad_norm": 66728.0390625,
|
| 371 |
+
"learning_rate": 8.843537414965988e-07,
|
| 372 |
+
"loss": 2.0219,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5408163265306123,
|
| 377 |
+
"grad_norm": 75688.15625,
|
| 378 |
+
"learning_rate": 9.013605442176871e-07,
|
| 379 |
+
"loss": 2.0165,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5510204081632653,
|
| 384 |
+
"grad_norm": 74948.90625,
|
| 385 |
+
"learning_rate": 9.183673469387756e-07,
|
| 386 |
+
"loss": 2.0228,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5612244897959183,
|
| 391 |
+
"grad_norm": 58538.5,
|
| 392 |
+
"learning_rate": 9.353741496598641e-07,
|
| 393 |
+
"loss": 2.0197,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5714285714285714,
|
| 398 |
+
"grad_norm": 58524.6796875,
|
| 399 |
+
"learning_rate": 9.523809523809525e-07,
|
| 400 |
+
"loss": 1.9988,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5816326530612245,
|
| 405 |
+
"grad_norm": 66490.1171875,
|
| 406 |
+
"learning_rate": 9.69387755102041e-07,
|
| 407 |
+
"loss": 1.9995,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5918367346938775,
|
| 412 |
+
"grad_norm": 76682.40625,
|
| 413 |
+
"learning_rate": 9.863945578231293e-07,
|
| 414 |
+
"loss": 2.0109,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6020408163265306,
|
| 419 |
+
"grad_norm": 76082.6953125,
|
| 420 |
+
"learning_rate": 1.0034013605442177e-06,
|
| 421 |
+
"loss": 1.9987,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6122448979591837,
|
| 426 |
+
"grad_norm": 75684.1796875,
|
| 427 |
+
"learning_rate": 1.0204081632653063e-06,
|
| 428 |
+
"loss": 1.9903,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6224489795918368,
|
| 433 |
+
"grad_norm": 108901.2890625,
|
| 434 |
+
"learning_rate": 1.0374149659863946e-06,
|
| 435 |
+
"loss": 1.9677,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6326530612244898,
|
| 440 |
+
"grad_norm": 85066.4375,
|
| 441 |
+
"learning_rate": 1.0544217687074832e-06,
|
| 442 |
+
"loss": 1.9941,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6428571428571429,
|
| 447 |
+
"grad_norm": 98574.0234375,
|
| 448 |
+
"learning_rate": 1.0714285714285714e-06,
|
| 449 |
+
"loss": 1.9778,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6530612244897959,
|
| 454 |
+
"grad_norm": 79184.9375,
|
| 455 |
+
"learning_rate": 1.08843537414966e-06,
|
| 456 |
+
"loss": 1.9723,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6632653061224489,
|
| 461 |
+
"grad_norm": 62203.5625,
|
| 462 |
+
"learning_rate": 1.1054421768707483e-06,
|
| 463 |
+
"loss": 1.9821,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.673469387755102,
|
| 468 |
+
"grad_norm": 69015.2734375,
|
| 469 |
+
"learning_rate": 1.122448979591837e-06,
|
| 470 |
+
"loss": 1.9708,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6836734693877551,
|
| 475 |
+
"grad_norm": 83259.71875,
|
| 476 |
+
"learning_rate": 1.1394557823129253e-06,
|
| 477 |
+
"loss": 1.9581,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6938775510204082,
|
| 482 |
+
"grad_norm": 93754.0546875,
|
| 483 |
+
"learning_rate": 1.1564625850340136e-06,
|
| 484 |
+
"loss": 1.9614,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7040816326530612,
|
| 489 |
+
"grad_norm": 97343.171875,
|
| 490 |
+
"learning_rate": 1.1734693877551022e-06,
|
| 491 |
+
"loss": 1.9372,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7142857142857143,
|
| 496 |
+
"grad_norm": 86450.984375,
|
| 497 |
+
"learning_rate": 1.1904761904761906e-06,
|
| 498 |
+
"loss": 1.9635,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7244897959183674,
|
| 503 |
+
"grad_norm": 109065.9140625,
|
| 504 |
+
"learning_rate": 1.207482993197279e-06,
|
| 505 |
+
"loss": 1.9432,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7346938775510204,
|
| 510 |
+
"grad_norm": 96424.5625,
|
| 511 |
+
"learning_rate": 1.2244897959183673e-06,
|
| 512 |
+
"loss": 1.9276,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7448979591836735,
|
| 517 |
+
"grad_norm": 91580.3671875,
|
| 518 |
+
"learning_rate": 1.241496598639456e-06,
|
| 519 |
+
"loss": 1.9075,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7551020408163265,
|
| 524 |
+
"grad_norm": 74389.96875,
|
| 525 |
+
"learning_rate": 1.2585034013605443e-06,
|
| 526 |
+
"loss": 1.9058,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7653061224489796,
|
| 531 |
+
"grad_norm": 114560.078125,
|
| 532 |
+
"learning_rate": 1.2755102040816329e-06,
|
| 533 |
+
"loss": 1.9247,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.7755102040816326,
|
| 538 |
+
"grad_norm": 104689.0078125,
|
| 539 |
+
"learning_rate": 1.2925170068027212e-06,
|
| 540 |
+
"loss": 1.9261,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7857142857142857,
|
| 545 |
+
"grad_norm": 115092.71875,
|
| 546 |
+
"learning_rate": 1.3095238095238096e-06,
|
| 547 |
+
"loss": 1.8644,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7959183673469388,
|
| 552 |
+
"grad_norm": 77653.03125,
|
| 553 |
+
"learning_rate": 1.3265306122448982e-06,
|
| 554 |
+
"loss": 1.8945,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8061224489795918,
|
| 559 |
+
"grad_norm": 97794.46875,
|
| 560 |
+
"learning_rate": 1.3435374149659866e-06,
|
| 561 |
+
"loss": 1.8651,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.8163265306122449,
|
| 566 |
+
"grad_norm": 123392.59375,
|
| 567 |
+
"learning_rate": 1.3605442176870751e-06,
|
| 568 |
+
"loss": 1.881,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.826530612244898,
|
| 573 |
+
"grad_norm": 78845.140625,
|
| 574 |
+
"learning_rate": 1.3775510204081633e-06,
|
| 575 |
+
"loss": 1.8774,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8367346938775511,
|
| 580 |
+
"grad_norm": 106911.1328125,
|
| 581 |
+
"learning_rate": 1.3945578231292517e-06,
|
| 582 |
+
"loss": 1.8784,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.8469387755102041,
|
| 587 |
+
"grad_norm": 97048.953125,
|
| 588 |
+
"learning_rate": 1.4115646258503403e-06,
|
| 589 |
+
"loss": 1.8848,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.8571428571428571,
|
| 594 |
+
"grad_norm": 112101.5859375,
|
| 595 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 596 |
+
"loss": 1.8932,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8673469387755102,
|
| 601 |
+
"grad_norm": 90785.6640625,
|
| 602 |
+
"learning_rate": 1.445578231292517e-06,
|
| 603 |
+
"loss": 1.8413,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8775510204081632,
|
| 608 |
+
"grad_norm": 150235.328125,
|
| 609 |
+
"learning_rate": 1.4625850340136056e-06,
|
| 610 |
+
"loss": 1.7915,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8877551020408163,
|
| 615 |
+
"grad_norm": 92524.75,
|
| 616 |
+
"learning_rate": 1.479591836734694e-06,
|
| 617 |
+
"loss": 1.8316,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.8979591836734694,
|
| 622 |
+
"grad_norm": 123792.203125,
|
| 623 |
+
"learning_rate": 1.4965986394557825e-06,
|
| 624 |
+
"loss": 1.7894,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9081632653061225,
|
| 629 |
+
"grad_norm": 105375.3671875,
|
| 630 |
+
"learning_rate": 1.513605442176871e-06,
|
| 631 |
+
"loss": 1.7826,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9183673469387755,
|
| 636 |
+
"grad_norm": 83843.8828125,
|
| 637 |
+
"learning_rate": 1.5306122448979593e-06,
|
| 638 |
+
"loss": 1.7905,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9285714285714286,
|
| 643 |
+
"grad_norm": 142571.53125,
|
| 644 |
+
"learning_rate": 1.5476190476190479e-06,
|
| 645 |
+
"loss": 1.8053,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9387755102040817,
|
| 650 |
+
"grad_norm": 119875.6328125,
|
| 651 |
+
"learning_rate": 1.5646258503401362e-06,
|
| 652 |
+
"loss": 1.8039,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9489795918367347,
|
| 657 |
+
"grad_norm": 112758.0390625,
|
| 658 |
+
"learning_rate": 1.5816326530612248e-06,
|
| 659 |
+
"loss": 1.7292,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9591836734693877,
|
| 664 |
+
"grad_norm": 125497.7890625,
|
| 665 |
+
"learning_rate": 1.5986394557823132e-06,
|
| 666 |
+
"loss": 1.7456,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9693877551020408,
|
| 671 |
+
"grad_norm": 110933.9609375,
|
| 672 |
+
"learning_rate": 1.6156462585034016e-06,
|
| 673 |
+
"loss": 1.7572,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9795918367346939,
|
| 678 |
+
"grad_norm": 149434.625,
|
| 679 |
+
"learning_rate": 1.6326530612244897e-06,
|
| 680 |
+
"loss": 1.7261,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9897959183673469,
|
| 685 |
+
"grad_norm": 149440.328125,
|
| 686 |
+
"learning_rate": 1.6496598639455783e-06,
|
| 687 |
+
"loss": 1.75,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.0,
|
| 692 |
+
"grad_norm": 189548.890625,
|
| 693 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 694 |
+
"loss": 1.7295,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.0,
|
| 699 |
+
"eval_accuracy": 0.4966465459423206,
|
| 700 |
+
"eval_loss": 1.6658564805984497,
|
| 701 |
+
"eval_runtime": 190.4552,
|
| 702 |
+
"eval_samples_per_second": 15.657,
|
| 703 |
+
"eval_steps_per_second": 0.656,
|
| 704 |
+
"step": 490
|
| 705 |
+
}
|
| 706 |
+
],
|
| 707 |
+
"logging_steps": 5,
|
| 708 |
+
"max_steps": 7350,
|
| 709 |
+
"num_input_tokens_seen": 0,
|
| 710 |
+
"num_train_epochs": 15,
|
| 711 |
+
"save_steps": 500,
|
| 712 |
+
"stateful_callbacks": {
|
| 713 |
+
"TrainerControl": {
|
| 714 |
+
"args": {
|
| 715 |
+
"should_epoch_stop": false,
|
| 716 |
+
"should_evaluate": false,
|
| 717 |
+
"should_log": false,
|
| 718 |
+
"should_save": true,
|
| 719 |
+
"should_training_stop": false
|
| 720 |
+
},
|
| 721 |
+
"attributes": {}
|
| 722 |
+
}
|
| 723 |
+
},
|
| 724 |
+
"total_flos": 1.49679069336576e+18,
|
| 725 |
+
"train_batch_size": 24,
|
| 726 |
+
"trial_name": null,
|
| 727 |
+
"trial_params": null
|
| 728 |
+
}
|
checkpoint-490/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c8330f8f1eea2830a3d99e288103b577fce3ec104205e265b5fe5492787aa5a
|
| 3 |
+
size 5176
|