Upload 6 files
Browse files
review_summary_generation/config.json
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/bart-large",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.1,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 12,
|
| 19 |
+
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"early_stopping": true,
|
| 22 |
+
"encoder_attention_heads": 16,
|
| 23 |
+
"encoder_ffn_dim": 4096,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 12,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"forced_bos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 2,
|
| 29 |
+
"gradient_checkpointing": false,
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "LABEL_0",
|
| 32 |
+
"1": "LABEL_1",
|
| 33 |
+
"2": "LABEL_2"
|
| 34 |
+
},
|
| 35 |
+
"init_std": 0.02,
|
| 36 |
+
"is_encoder_decoder": true,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"LABEL_0": 0,
|
| 39 |
+
"LABEL_1": 1,
|
| 40 |
+
"LABEL_2": 2
|
| 41 |
+
},
|
| 42 |
+
"max_position_embeddings": 1024,
|
| 43 |
+
"model_type": "bart",
|
| 44 |
+
"no_repeat_ngram_size": 3,
|
| 45 |
+
"normalize_before": false,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"num_hidden_layers": 12,
|
| 48 |
+
"pad_token_id": 1,
|
| 49 |
+
"scale_embedding": false,
|
| 50 |
+
"task_specific_params": {
|
| 51 |
+
"summarization": {
|
| 52 |
+
"length_penalty": 1.0,
|
| 53 |
+
"max_length": 128,
|
| 54 |
+
"min_length": 12,
|
| 55 |
+
"num_beams": 4
|
| 56 |
+
},
|
| 57 |
+
"summarization_cnn": {
|
| 58 |
+
"length_penalty": 2.0,
|
| 59 |
+
"max_length": 142,
|
| 60 |
+
"min_length": 56,
|
| 61 |
+
"num_beams": 4
|
| 62 |
+
},
|
| 63 |
+
"summarization_xsum": {
|
| 64 |
+
"length_penalty": 1.0,
|
| 65 |
+
"max_length": 62,
|
| 66 |
+
"min_length": 11,
|
| 67 |
+
"num_beams": 6
|
| 68 |
+
}
|
| 69 |
+
},
|
| 70 |
+
"torch_dtype": "float32",
|
| 71 |
+
"transformers_version": "4.42.4",
|
| 72 |
+
"use_cache": true,
|
| 73 |
+
"vocab_size": 50265
|
| 74 |
+
}
|
review_summary_generation/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"no_repeat_ngram_size": 3,
|
| 10 |
+
"num_beams": 4,
|
| 11 |
+
"pad_token_id": 1,
|
| 12 |
+
"transformers_version": "4.42.4"
|
| 13 |
+
}
|
review_summary_generation/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56e45de19d7cfa0eee5f9aa408806978a5abb80e9ae7def2827d0162440d65eb
|
| 3 |
+
size 14244
|
review_summary_generation/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e7ac0ebed9c55dfbf97702b1b950706ef513614e6bc86c950547b9b85977dfc
|
| 3 |
+
size 1064
|
review_summary_generation/trainer_state.json
ADDED
|
@@ -0,0 +1,1107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.5866928100585938,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-500",
|
| 4 |
+
"epoch": 3.1982942430703627,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.021321961620469083,
|
| 13 |
+
"grad_norm": 17.069480895996094,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"loss": 5.4109,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.042643923240938165,
|
| 20 |
+
"grad_norm": 14.406121253967285,
|
| 21 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 22 |
+
"loss": 5.1227,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06396588486140725,
|
| 27 |
+
"grad_norm": 15.006143569946289,
|
| 28 |
+
"learning_rate": 3e-06,
|
| 29 |
+
"loss": 4.4532,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08528784648187633,
|
| 34 |
+
"grad_norm": 104.21044158935547,
|
| 35 |
+
"learning_rate": 4.000000000000001e-06,
|
| 36 |
+
"loss": 4.0782,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10660980810234541,
|
| 41 |
+
"grad_norm": 11.47994613647461,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 4.9843,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1279317697228145,
|
| 48 |
+
"grad_norm": 10.76855754852295,
|
| 49 |
+
"learning_rate": 6e-06,
|
| 50 |
+
"loss": 4.047,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14925373134328357,
|
| 55 |
+
"grad_norm": 9.982112884521484,
|
| 56 |
+
"learning_rate": 7.000000000000001e-06,
|
| 57 |
+
"loss": 3.7253,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.17057569296375266,
|
| 62 |
+
"grad_norm": 10.03783893585205,
|
| 63 |
+
"learning_rate": 8.000000000000001e-06,
|
| 64 |
+
"loss": 4.0644,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.19189765458422176,
|
| 69 |
+
"grad_norm": 10.862977981567383,
|
| 70 |
+
"learning_rate": 9e-06,
|
| 71 |
+
"loss": 3.9467,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.21321961620469082,
|
| 76 |
+
"grad_norm": 12.349374771118164,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 4.0981,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2345415778251599,
|
| 83 |
+
"grad_norm": 10.292299270629883,
|
| 84 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 85 |
+
"loss": 3.5753,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.255863539445629,
|
| 90 |
+
"grad_norm": 10.439337730407715,
|
| 91 |
+
"learning_rate": 1.2e-05,
|
| 92 |
+
"loss": 3.66,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2771855010660981,
|
| 97 |
+
"grad_norm": 9.760316848754883,
|
| 98 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 99 |
+
"loss": 3.702,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.29850746268656714,
|
| 104 |
+
"grad_norm": 10.415947914123535,
|
| 105 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 106 |
+
"loss": 4.0965,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.31982942430703626,
|
| 111 |
+
"grad_norm": 10.626498222351074,
|
| 112 |
+
"learning_rate": 1.5e-05,
|
| 113 |
+
"loss": 3.9496,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3411513859275053,
|
| 118 |
+
"grad_norm": 10.134171485900879,
|
| 119 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 120 |
+
"loss": 3.9231,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3624733475479744,
|
| 125 |
+
"grad_norm": 11.33432674407959,
|
| 126 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 127 |
+
"loss": 3.409,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3837953091684435,
|
| 132 |
+
"grad_norm": 12.020699501037598,
|
| 133 |
+
"learning_rate": 1.8e-05,
|
| 134 |
+
"loss": 3.5327,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4051172707889126,
|
| 139 |
+
"grad_norm": 10.670236587524414,
|
| 140 |
+
"learning_rate": 1.9e-05,
|
| 141 |
+
"loss": 3.5801,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.42643923240938164,
|
| 146 |
+
"grad_norm": 11.246965408325195,
|
| 147 |
+
"learning_rate": 2e-05,
|
| 148 |
+
"loss": 3.5045,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.44776119402985076,
|
| 153 |
+
"grad_norm": 12.235991477966309,
|
| 154 |
+
"learning_rate": 2.1e-05,
|
| 155 |
+
"loss": 3.9952,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4690831556503198,
|
| 160 |
+
"grad_norm": 17.548768997192383,
|
| 161 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 162 |
+
"loss": 3.3532,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4904051172707889,
|
| 167 |
+
"grad_norm": 19.35828971862793,
|
| 168 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 169 |
+
"loss": 3.3175,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.511727078891258,
|
| 174 |
+
"grad_norm": 10.685091972351074,
|
| 175 |
+
"learning_rate": 2.4e-05,
|
| 176 |
+
"loss": 3.2104,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5330490405117271,
|
| 181 |
+
"grad_norm": 10.596184730529785,
|
| 182 |
+
"learning_rate": 2.5e-05,
|
| 183 |
+
"loss": 3.412,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5543710021321961,
|
| 188 |
+
"grad_norm": 11.168882369995117,
|
| 189 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 190 |
+
"loss": 3.2391,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5756929637526652,
|
| 195 |
+
"grad_norm": 10.356324195861816,
|
| 196 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 197 |
+
"loss": 3.1957,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5970149253731343,
|
| 202 |
+
"grad_norm": 8.388641357421875,
|
| 203 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 204 |
+
"loss": 3.0793,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6183368869936035,
|
| 209 |
+
"grad_norm": 8.973965644836426,
|
| 210 |
+
"learning_rate": 2.9e-05,
|
| 211 |
+
"loss": 3.319,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6396588486140725,
|
| 216 |
+
"grad_norm": 12.515954971313477,
|
| 217 |
+
"learning_rate": 3e-05,
|
| 218 |
+
"loss": 2.9707,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6609808102345416,
|
| 223 |
+
"grad_norm": 10.036434173583984,
|
| 224 |
+
"learning_rate": 3.1e-05,
|
| 225 |
+
"loss": 2.834,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6823027718550106,
|
| 230 |
+
"grad_norm": 10.861001014709473,
|
| 231 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 232 |
+
"loss": 2.8357,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7036247334754797,
|
| 237 |
+
"grad_norm": 8.849268913269043,
|
| 238 |
+
"learning_rate": 3.3e-05,
|
| 239 |
+
"loss": 3.0412,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7249466950959488,
|
| 244 |
+
"grad_norm": 9.351764678955078,
|
| 245 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 246 |
+
"loss": 2.8768,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.746268656716418,
|
| 251 |
+
"grad_norm": 9.140671730041504,
|
| 252 |
+
"learning_rate": 3.5e-05,
|
| 253 |
+
"loss": 3.0797,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.767590618336887,
|
| 258 |
+
"grad_norm": 9.828067779541016,
|
| 259 |
+
"learning_rate": 3.6e-05,
|
| 260 |
+
"loss": 2.8023,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7889125799573561,
|
| 265 |
+
"grad_norm": 29.474517822265625,
|
| 266 |
+
"learning_rate": 3.7e-05,
|
| 267 |
+
"loss": 2.9934,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8102345415778252,
|
| 272 |
+
"grad_norm": 9.97517204284668,
|
| 273 |
+
"learning_rate": 3.8e-05,
|
| 274 |
+
"loss": 3.0032,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8315565031982942,
|
| 279 |
+
"grad_norm": 16.823415756225586,
|
| 280 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 281 |
+
"loss": 3.0508,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8528784648187633,
|
| 286 |
+
"grad_norm": 7.544747352600098,
|
| 287 |
+
"learning_rate": 4e-05,
|
| 288 |
+
"loss": 2.9423,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8742004264392325,
|
| 293 |
+
"grad_norm": 7.29916524887085,
|
| 294 |
+
"learning_rate": 4.1e-05,
|
| 295 |
+
"loss": 2.6364,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8955223880597015,
|
| 300 |
+
"grad_norm": 7.249739646911621,
|
| 301 |
+
"learning_rate": 4.2e-05,
|
| 302 |
+
"loss": 2.7285,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9168443496801706,
|
| 307 |
+
"grad_norm": 8.090813636779785,
|
| 308 |
+
"learning_rate": 4.3e-05,
|
| 309 |
+
"loss": 2.7175,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9381663113006397,
|
| 314 |
+
"grad_norm": 10.216124534606934,
|
| 315 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 316 |
+
"loss": 3.182,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9594882729211087,
|
| 321 |
+
"grad_norm": 6.813947677612305,
|
| 322 |
+
"learning_rate": 4.5e-05,
|
| 323 |
+
"loss": 2.7877,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9808102345415778,
|
| 328 |
+
"grad_norm": 13.773616790771484,
|
| 329 |
+
"learning_rate": 4.600000000000001e-05,
|
| 330 |
+
"loss": 2.7575,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.0021321961620469,
|
| 335 |
+
"grad_norm": 10.226593017578125,
|
| 336 |
+
"learning_rate": 4.7e-05,
|
| 337 |
+
"loss": 2.5982,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.023454157782516,
|
| 342 |
+
"grad_norm": 7.802149295806885,
|
| 343 |
+
"learning_rate": 4.8e-05,
|
| 344 |
+
"loss": 2.6878,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.044776119402985,
|
| 349 |
+
"grad_norm": 8.879491806030273,
|
| 350 |
+
"learning_rate": 4.9e-05,
|
| 351 |
+
"loss": 2.6006,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.0660980810234542,
|
| 356 |
+
"grad_norm": 7.746878623962402,
|
| 357 |
+
"learning_rate": 5e-05,
|
| 358 |
+
"loss": 2.7281,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.0660980810234542,
|
| 363 |
+
"eval_loss": 2.5866928100585938,
|
| 364 |
+
"eval_runtime": 17.7541,
|
| 365 |
+
"eval_samples_per_second": 11.321,
|
| 366 |
+
"eval_steps_per_second": 5.689,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0874200426439233,
|
| 371 |
+
"grad_norm": 6.854872703552246,
|
| 372 |
+
"learning_rate": 4.988066825775656e-05,
|
| 373 |
+
"loss": 2.5271,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.1087420042643923,
|
| 378 |
+
"grad_norm": 14.891672134399414,
|
| 379 |
+
"learning_rate": 4.976133651551313e-05,
|
| 380 |
+
"loss": 2.5794,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.1300639658848615,
|
| 385 |
+
"grad_norm": 9.460530281066895,
|
| 386 |
+
"learning_rate": 4.964200477326969e-05,
|
| 387 |
+
"loss": 2.863,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.1513859275053304,
|
| 392 |
+
"grad_norm": 8.543593406677246,
|
| 393 |
+
"learning_rate": 4.952267303102625e-05,
|
| 394 |
+
"loss": 2.8392,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.1727078891257996,
|
| 399 |
+
"grad_norm": 7.970772743225098,
|
| 400 |
+
"learning_rate": 4.940334128878282e-05,
|
| 401 |
+
"loss": 2.4553,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.1940298507462686,
|
| 406 |
+
"grad_norm": 9.082477569580078,
|
| 407 |
+
"learning_rate": 4.9284009546539385e-05,
|
| 408 |
+
"loss": 2.594,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.2153518123667377,
|
| 413 |
+
"grad_norm": 7.294309139251709,
|
| 414 |
+
"learning_rate": 4.9164677804295945e-05,
|
| 415 |
+
"loss": 2.3664,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.236673773987207,
|
| 420 |
+
"grad_norm": 9.168344497680664,
|
| 421 |
+
"learning_rate": 4.904534606205251e-05,
|
| 422 |
+
"loss": 2.7231,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.2579957356076759,
|
| 427 |
+
"grad_norm": 7.526830196380615,
|
| 428 |
+
"learning_rate": 4.892601431980907e-05,
|
| 429 |
+
"loss": 2.3026,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.279317697228145,
|
| 434 |
+
"grad_norm": 10.8075590133667,
|
| 435 |
+
"learning_rate": 4.880668257756563e-05,
|
| 436 |
+
"loss": 2.5006,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.3006396588486142,
|
| 441 |
+
"grad_norm": 11.046674728393555,
|
| 442 |
+
"learning_rate": 4.8687350835322196e-05,
|
| 443 |
+
"loss": 2.6506,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.3219616204690832,
|
| 448 |
+
"grad_norm": 13.209379196166992,
|
| 449 |
+
"learning_rate": 4.856801909307876e-05,
|
| 450 |
+
"loss": 2.9135,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.3432835820895521,
|
| 455 |
+
"grad_norm": 15.499232292175293,
|
| 456 |
+
"learning_rate": 4.844868735083533e-05,
|
| 457 |
+
"loss": 2.8046,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.3646055437100213,
|
| 462 |
+
"grad_norm": 8.003043174743652,
|
| 463 |
+
"learning_rate": 4.832935560859189e-05,
|
| 464 |
+
"loss": 2.3684,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.3859275053304905,
|
| 469 |
+
"grad_norm": 11.293377876281738,
|
| 470 |
+
"learning_rate": 4.821002386634845e-05,
|
| 471 |
+
"loss": 2.5306,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.4072494669509594,
|
| 476 |
+
"grad_norm": 6.7576398849487305,
|
| 477 |
+
"learning_rate": 4.809069212410501e-05,
|
| 478 |
+
"loss": 2.3224,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.4285714285714286,
|
| 483 |
+
"grad_norm": 9.321499824523926,
|
| 484 |
+
"learning_rate": 4.797136038186158e-05,
|
| 485 |
+
"loss": 2.8741,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.4498933901918978,
|
| 490 |
+
"grad_norm": 6.4436421394348145,
|
| 491 |
+
"learning_rate": 4.785202863961814e-05,
|
| 492 |
+
"loss": 2.3508,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.4712153518123667,
|
| 497 |
+
"grad_norm": 11.9252347946167,
|
| 498 |
+
"learning_rate": 4.7732696897374704e-05,
|
| 499 |
+
"loss": 2.528,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.4925373134328357,
|
| 504 |
+
"grad_norm": 6.628900051116943,
|
| 505 |
+
"learning_rate": 4.761336515513127e-05,
|
| 506 |
+
"loss": 2.455,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.5138592750533049,
|
| 511 |
+
"grad_norm": 9.39482593536377,
|
| 512 |
+
"learning_rate": 4.749403341288783e-05,
|
| 513 |
+
"loss": 2.3555,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.535181236673774,
|
| 518 |
+
"grad_norm": 10.35986042022705,
|
| 519 |
+
"learning_rate": 4.7374701670644396e-05,
|
| 520 |
+
"loss": 2.5701,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.556503198294243,
|
| 525 |
+
"grad_norm": 6.811497688293457,
|
| 526 |
+
"learning_rate": 4.7255369928400955e-05,
|
| 527 |
+
"loss": 2.7697,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.5778251599147122,
|
| 532 |
+
"grad_norm": 8.669289588928223,
|
| 533 |
+
"learning_rate": 4.713603818615752e-05,
|
| 534 |
+
"loss": 2.4659,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.5991471215351813,
|
| 539 |
+
"grad_norm": 7.620774745941162,
|
| 540 |
+
"learning_rate": 4.701670644391408e-05,
|
| 541 |
+
"loss": 2.5101,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.6204690831556503,
|
| 546 |
+
"grad_norm": 10.02075481414795,
|
| 547 |
+
"learning_rate": 4.6897374701670647e-05,
|
| 548 |
+
"loss": 2.4799,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.6417910447761193,
|
| 553 |
+
"grad_norm": 11.186421394348145,
|
| 554 |
+
"learning_rate": 4.677804295942721e-05,
|
| 555 |
+
"loss": 2.5446,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.6631130063965884,
|
| 560 |
+
"grad_norm": 8.406248092651367,
|
| 561 |
+
"learning_rate": 4.665871121718377e-05,
|
| 562 |
+
"loss": 2.2934,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.6844349680170576,
|
| 567 |
+
"grad_norm": 12.06204605102539,
|
| 568 |
+
"learning_rate": 4.653937947494034e-05,
|
| 569 |
+
"loss": 2.5549,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.7057569296375266,
|
| 574 |
+
"grad_norm": 13.223321914672852,
|
| 575 |
+
"learning_rate": 4.64200477326969e-05,
|
| 576 |
+
"loss": 2.7932,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.7270788912579957,
|
| 581 |
+
"grad_norm": 9.57943058013916,
|
| 582 |
+
"learning_rate": 4.6300715990453463e-05,
|
| 583 |
+
"loss": 2.7192,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.748400852878465,
|
| 588 |
+
"grad_norm": 9.070585250854492,
|
| 589 |
+
"learning_rate": 4.618138424821002e-05,
|
| 590 |
+
"loss": 2.5365,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.7697228144989339,
|
| 595 |
+
"grad_norm": 8.995747566223145,
|
| 596 |
+
"learning_rate": 4.606205250596659e-05,
|
| 597 |
+
"loss": 2.7967,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.7910447761194028,
|
| 602 |
+
"grad_norm": 10.535736083984375,
|
| 603 |
+
"learning_rate": 4.594272076372315e-05,
|
| 604 |
+
"loss": 2.6245,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.8123667377398722,
|
| 609 |
+
"grad_norm": 9.777527809143066,
|
| 610 |
+
"learning_rate": 4.582338902147972e-05,
|
| 611 |
+
"loss": 2.7633,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.8336886993603412,
|
| 616 |
+
"grad_norm": 7.4152069091796875,
|
| 617 |
+
"learning_rate": 4.570405727923628e-05,
|
| 618 |
+
"loss": 2.539,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.8550106609808101,
|
| 623 |
+
"grad_norm": 9.038530349731445,
|
| 624 |
+
"learning_rate": 4.5584725536992847e-05,
|
| 625 |
+
"loss": 2.7307,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.8763326226012793,
|
| 630 |
+
"grad_norm": 6.49500036239624,
|
| 631 |
+
"learning_rate": 4.5465393794749406e-05,
|
| 632 |
+
"loss": 2.4288,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.8976545842217485,
|
| 637 |
+
"grad_norm": 8.275555610656738,
|
| 638 |
+
"learning_rate": 4.5346062052505965e-05,
|
| 639 |
+
"loss": 2.6612,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.9189765458422174,
|
| 644 |
+
"grad_norm": 7.9432373046875,
|
| 645 |
+
"learning_rate": 4.522673031026253e-05,
|
| 646 |
+
"loss": 2.4942,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.9402985074626866,
|
| 651 |
+
"grad_norm": 8.501258850097656,
|
| 652 |
+
"learning_rate": 4.510739856801909e-05,
|
| 653 |
+
"loss": 2.3821,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.9616204690831558,
|
| 658 |
+
"grad_norm": 8.605585098266602,
|
| 659 |
+
"learning_rate": 4.498806682577566e-05,
|
| 660 |
+
"loss": 2.5933,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.9829424307036247,
|
| 665 |
+
"grad_norm": 8.423508644104004,
|
| 666 |
+
"learning_rate": 4.486873508353222e-05,
|
| 667 |
+
"loss": 2.6085,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.0042643923240937,
|
| 672 |
+
"grad_norm": 8.394946098327637,
|
| 673 |
+
"learning_rate": 4.474940334128879e-05,
|
| 674 |
+
"loss": 2.452,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.025586353944563,
|
| 679 |
+
"grad_norm": 6.4166436195373535,
|
| 680 |
+
"learning_rate": 4.463007159904535e-05,
|
| 681 |
+
"loss": 1.936,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.046908315565032,
|
| 686 |
+
"grad_norm": 9.592107772827148,
|
| 687 |
+
"learning_rate": 4.4510739856801914e-05,
|
| 688 |
+
"loss": 2.1497,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.068230277185501,
|
| 693 |
+
"grad_norm": 5.840519428253174,
|
| 694 |
+
"learning_rate": 4.4391408114558474e-05,
|
| 695 |
+
"loss": 1.8572,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.08955223880597,
|
| 700 |
+
"grad_norm": 8.546823501586914,
|
| 701 |
+
"learning_rate": 4.427207637231503e-05,
|
| 702 |
+
"loss": 2.2502,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.1108742004264394,
|
| 707 |
+
"grad_norm": 9.255841255187988,
|
| 708 |
+
"learning_rate": 4.41527446300716e-05,
|
| 709 |
+
"loss": 1.9999,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.1321961620469083,
|
| 714 |
+
"grad_norm": 8.582698822021484,
|
| 715 |
+
"learning_rate": 4.4033412887828165e-05,
|
| 716 |
+
"loss": 1.9693,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.1321961620469083,
|
| 721 |
+
"eval_loss": 2.7102935314178467,
|
| 722 |
+
"eval_runtime": 17.5751,
|
| 723 |
+
"eval_samples_per_second": 11.437,
|
| 724 |
+
"eval_steps_per_second": 5.747,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.1535181236673773,
|
| 729 |
+
"grad_norm": 8.656256675720215,
|
| 730 |
+
"learning_rate": 4.391408114558473e-05,
|
| 731 |
+
"loss": 1.8755,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.1748400852878467,
|
| 736 |
+
"grad_norm": 8.566282272338867,
|
| 737 |
+
"learning_rate": 4.379474940334129e-05,
|
| 738 |
+
"loss": 2.132,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.1961620469083156,
|
| 743 |
+
"grad_norm": 8.829427719116211,
|
| 744 |
+
"learning_rate": 4.367541766109786e-05,
|
| 745 |
+
"loss": 2.1197,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.2174840085287846,
|
| 750 |
+
"grad_norm": 7.132998466491699,
|
| 751 |
+
"learning_rate": 4.3556085918854416e-05,
|
| 752 |
+
"loss": 2.135,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.2388059701492535,
|
| 757 |
+
"grad_norm": 11.001921653747559,
|
| 758 |
+
"learning_rate": 4.343675417661098e-05,
|
| 759 |
+
"loss": 2.1064,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.260127931769723,
|
| 764 |
+
"grad_norm": 10.78303050994873,
|
| 765 |
+
"learning_rate": 4.331742243436754e-05,
|
| 766 |
+
"loss": 1.9293,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.281449893390192,
|
| 771 |
+
"grad_norm": 6.237246036529541,
|
| 772 |
+
"learning_rate": 4.319809069212411e-05,
|
| 773 |
+
"loss": 1.8999,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.302771855010661,
|
| 778 |
+
"grad_norm": 6.525657653808594,
|
| 779 |
+
"learning_rate": 4.3078758949880674e-05,
|
| 780 |
+
"loss": 2.1728,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.3240938166311302,
|
| 785 |
+
"grad_norm": 8.797844886779785,
|
| 786 |
+
"learning_rate": 4.295942720763723e-05,
|
| 787 |
+
"loss": 1.8529,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.345415778251599,
|
| 792 |
+
"grad_norm": 7.462911605834961,
|
| 793 |
+
"learning_rate": 4.28400954653938e-05,
|
| 794 |
+
"loss": 1.8816,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.366737739872068,
|
| 799 |
+
"grad_norm": 9.956457138061523,
|
| 800 |
+
"learning_rate": 4.272076372315036e-05,
|
| 801 |
+
"loss": 2.2707,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.388059701492537,
|
| 806 |
+
"grad_norm": 13.278641700744629,
|
| 807 |
+
"learning_rate": 4.2601431980906925e-05,
|
| 808 |
+
"loss": 1.9471,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.4093816631130065,
|
| 813 |
+
"grad_norm": 14.353861808776855,
|
| 814 |
+
"learning_rate": 4.2482100238663484e-05,
|
| 815 |
+
"loss": 2.0054,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.4307036247334755,
|
| 820 |
+
"grad_norm": 9.295260429382324,
|
| 821 |
+
"learning_rate": 4.236276849642005e-05,
|
| 822 |
+
"loss": 2.2011,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.4520255863539444,
|
| 827 |
+
"grad_norm": 7.920892715454102,
|
| 828 |
+
"learning_rate": 4.2243436754176616e-05,
|
| 829 |
+
"loss": 2.0244,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.473347547974414,
|
| 834 |
+
"grad_norm": 6.263861179351807,
|
| 835 |
+
"learning_rate": 4.2124105011933175e-05,
|
| 836 |
+
"loss": 1.9248,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.4946695095948828,
|
| 841 |
+
"grad_norm": 6.800403118133545,
|
| 842 |
+
"learning_rate": 4.200477326968974e-05,
|
| 843 |
+
"loss": 2.0803,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.5159914712153517,
|
| 848 |
+
"grad_norm": 10.121644973754883,
|
| 849 |
+
"learning_rate": 4.18854415274463e-05,
|
| 850 |
+
"loss": 1.8253,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.5373134328358207,
|
| 855 |
+
"grad_norm": 9.465630531311035,
|
| 856 |
+
"learning_rate": 4.176610978520287e-05,
|
| 857 |
+
"loss": 2.0638,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.55863539445629,
|
| 862 |
+
"grad_norm": 8.128814697265625,
|
| 863 |
+
"learning_rate": 4.1646778042959426e-05,
|
| 864 |
+
"loss": 2.0926,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.579957356076759,
|
| 869 |
+
"grad_norm": 8.691697120666504,
|
| 870 |
+
"learning_rate": 4.152744630071599e-05,
|
| 871 |
+
"loss": 1.7781,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.6012793176972284,
|
| 876 |
+
"grad_norm": 8.142535209655762,
|
| 877 |
+
"learning_rate": 4.140811455847255e-05,
|
| 878 |
+
"loss": 2.0987,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.6226012793176974,
|
| 883 |
+
"grad_norm": 7.992061138153076,
|
| 884 |
+
"learning_rate": 4.1288782816229125e-05,
|
| 885 |
+
"loss": 2.1963,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 2.6439232409381663,
|
| 890 |
+
"grad_norm": 6.09903621673584,
|
| 891 |
+
"learning_rate": 4.1169451073985684e-05,
|
| 892 |
+
"loss": 1.753,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.6652452025586353,
|
| 897 |
+
"grad_norm": 7.56857967376709,
|
| 898 |
+
"learning_rate": 4.105011933174224e-05,
|
| 899 |
+
"loss": 2.1974,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.6865671641791042,
|
| 904 |
+
"grad_norm": 7.062485218048096,
|
| 905 |
+
"learning_rate": 4.093078758949881e-05,
|
| 906 |
+
"loss": 1.7141,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.7078891257995736,
|
| 911 |
+
"grad_norm": 7.561300277709961,
|
| 912 |
+
"learning_rate": 4.081145584725537e-05,
|
| 913 |
+
"loss": 2.0716,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.7292110874200426,
|
| 918 |
+
"grad_norm": 7.001197814941406,
|
| 919 |
+
"learning_rate": 4.0692124105011935e-05,
|
| 920 |
+
"loss": 2.102,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.750533049040512,
|
| 925 |
+
"grad_norm": 6.451600074768066,
|
| 926 |
+
"learning_rate": 4.0572792362768494e-05,
|
| 927 |
+
"loss": 2.0389,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.771855010660981,
|
| 932 |
+
"grad_norm": 6.065812110900879,
|
| 933 |
+
"learning_rate": 4.045346062052506e-05,
|
| 934 |
+
"loss": 1.9661,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.79317697228145,
|
| 939 |
+
"grad_norm": 12.280841827392578,
|
| 940 |
+
"learning_rate": 4.0334128878281626e-05,
|
| 941 |
+
"loss": 2.1364,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.814498933901919,
|
| 946 |
+
"grad_norm": 6.758681297302246,
|
| 947 |
+
"learning_rate": 4.021479713603819e-05,
|
| 948 |
+
"loss": 2.1638,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.835820895522388,
|
| 953 |
+
"grad_norm": 10.302852630615234,
|
| 954 |
+
"learning_rate": 4.009546539379475e-05,
|
| 955 |
+
"loss": 2.037,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.857142857142857,
|
| 960 |
+
"grad_norm": 8.164216041564941,
|
| 961 |
+
"learning_rate": 3.997613365155131e-05,
|
| 962 |
+
"loss": 1.9582,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.878464818763326,
|
| 967 |
+
"grad_norm": 7.020501136779785,
|
| 968 |
+
"learning_rate": 3.985680190930788e-05,
|
| 969 |
+
"loss": 2.2448,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 2.8997867803837956,
|
| 974 |
+
"grad_norm": 8.621040344238281,
|
| 975 |
+
"learning_rate": 3.9737470167064437e-05,
|
| 976 |
+
"loss": 1.9945,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.9211087420042645,
|
| 981 |
+
"grad_norm": 8.973491668701172,
|
| 982 |
+
"learning_rate": 3.9618138424821e-05,
|
| 983 |
+
"loss": 2.1009,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.9424307036247335,
|
| 988 |
+
"grad_norm": 12.35886001586914,
|
| 989 |
+
"learning_rate": 3.949880668257757e-05,
|
| 990 |
+
"loss": 2.2309,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.9637526652452024,
|
| 995 |
+
"grad_norm": 7.0847296714782715,
|
| 996 |
+
"learning_rate": 3.9379474940334135e-05,
|
| 997 |
+
"loss": 2.1275,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.9850746268656714,
|
| 1002 |
+
"grad_norm": 8.902274131774902,
|
| 1003 |
+
"learning_rate": 3.9260143198090694e-05,
|
| 1004 |
+
"loss": 2.146,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 3.0063965884861408,
|
| 1009 |
+
"grad_norm": 7.1567463874816895,
|
| 1010 |
+
"learning_rate": 3.914081145584726e-05,
|
| 1011 |
+
"loss": 2.1079,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 3.0277185501066097,
|
| 1016 |
+
"grad_norm": 10.294557571411133,
|
| 1017 |
+
"learning_rate": 3.902147971360382e-05,
|
| 1018 |
+
"loss": 1.6892,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 3.0490405117270787,
|
| 1023 |
+
"grad_norm": 11.8173246383667,
|
| 1024 |
+
"learning_rate": 3.8902147971360386e-05,
|
| 1025 |
+
"loss": 1.7492,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 3.070362473347548,
|
| 1030 |
+
"grad_norm": 7.909124851226807,
|
| 1031 |
+
"learning_rate": 3.8782816229116945e-05,
|
| 1032 |
+
"loss": 1.5643,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 3.091684434968017,
|
| 1037 |
+
"grad_norm": 7.754435062408447,
|
| 1038 |
+
"learning_rate": 3.866348448687351e-05,
|
| 1039 |
+
"loss": 1.6751,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.113006396588486,
|
| 1044 |
+
"grad_norm": 9.501265525817871,
|
| 1045 |
+
"learning_rate": 3.854415274463008e-05,
|
| 1046 |
+
"loss": 1.5605,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 3.1343283582089554,
|
| 1051 |
+
"grad_norm": 7.292548656463623,
|
| 1052 |
+
"learning_rate": 3.8424821002386637e-05,
|
| 1053 |
+
"loss": 1.8255,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 3.1556503198294243,
|
| 1058 |
+
"grad_norm": 9.267513275146484,
|
| 1059 |
+
"learning_rate": 3.83054892601432e-05,
|
| 1060 |
+
"loss": 1.5421,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 3.1769722814498933,
|
| 1065 |
+
"grad_norm": 9.507182121276855,
|
| 1066 |
+
"learning_rate": 3.818615751789976e-05,
|
| 1067 |
+
"loss": 1.6099,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 3.1982942430703627,
|
| 1072 |
+
"grad_norm": 7.819943904876709,
|
| 1073 |
+
"learning_rate": 3.806682577565633e-05,
|
| 1074 |
+
"loss": 1.4942,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 3.1982942430703627,
|
| 1079 |
+
"eval_loss": 2.8242695331573486,
|
| 1080 |
+
"eval_runtime": 17.5083,
|
| 1081 |
+
"eval_samples_per_second": 11.48,
|
| 1082 |
+
"eval_steps_per_second": 5.769,
|
| 1083 |
+
"step": 1500
|
| 1084 |
+
}
|
| 1085 |
+
],
|
| 1086 |
+
"logging_steps": 10,
|
| 1087 |
+
"max_steps": 4690,
|
| 1088 |
+
"num_input_tokens_seen": 0,
|
| 1089 |
+
"num_train_epochs": 10,
|
| 1090 |
+
"save_steps": 500,
|
| 1091 |
+
"stateful_callbacks": {
|
| 1092 |
+
"TrainerControl": {
|
| 1093 |
+
"args": {
|
| 1094 |
+
"should_epoch_stop": false,
|
| 1095 |
+
"should_evaluate": false,
|
| 1096 |
+
"should_log": false,
|
| 1097 |
+
"should_save": true,
|
| 1098 |
+
"should_training_stop": false
|
| 1099 |
+
},
|
| 1100 |
+
"attributes": {}
|
| 1101 |
+
}
|
| 1102 |
+
},
|
| 1103 |
+
"total_flos": 3247406246264832.0,
|
| 1104 |
+
"train_batch_size": 2,
|
| 1105 |
+
"trial_name": null,
|
| 1106 |
+
"trial_params": null
|
| 1107 |
+
}
|
review_summary_generation/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25727f64e02e6ac432d9d980059bae044dd8e449837e7c0828cdff67dc502d88
|
| 3 |
+
size 5048
|