Upload folder using huggingface_hub
Browse files- added_tokens.json +7 -0
- config.json +113 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +27 -0
- tokenizer.json +0 -0
- tokenizer_config.json +68 -0
- trainer_state.json +2145 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<<LABEL>>": 151646,
|
| 3 |
+
"<<SEP>>": 151647,
|
| 4 |
+
"<|endoftext|>": 151643,
|
| 5 |
+
"<|im_end|>": 151645,
|
| 6 |
+
"<|im_start|>": 151644
|
| 7 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "models/qwen/base",
|
| 3 |
+
"architecture_type": "uni-encoder",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GLiClassModel"
|
| 6 |
+
],
|
| 7 |
+
"class_token_index": 151646,
|
| 8 |
+
"contrastive_loss_coef": 0.03,
|
| 9 |
+
"embed_class_token": true,
|
| 10 |
+
"encoder_config": {
|
| 11 |
+
"_name_or_path": "knowledgator/Qwen-encoder-0.5B",
|
| 12 |
+
"add_cross_attention": false,
|
| 13 |
+
"architectures": [
|
| 14 |
+
"Qwen2ForCausalLM"
|
| 15 |
+
],
|
| 16 |
+
"attention_dropout": 0.0,
|
| 17 |
+
"bad_words_ids": null,
|
| 18 |
+
"begin_suppress_tokens": null,
|
| 19 |
+
"bos_token_id": 151643,
|
| 20 |
+
"chunk_size_feed_forward": 0,
|
| 21 |
+
"cross_attention_hidden_size": null,
|
| 22 |
+
"decoder_start_token_id": null,
|
| 23 |
+
"diversity_penalty": 0.0,
|
| 24 |
+
"do_sample": false,
|
| 25 |
+
"early_stopping": false,
|
| 26 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 27 |
+
"eos_token_id": 151643,
|
| 28 |
+
"exponential_decay_length_penalty": null,
|
| 29 |
+
"finetuning_task": null,
|
| 30 |
+
"forced_bos_token_id": null,
|
| 31 |
+
"forced_eos_token_id": null,
|
| 32 |
+
"hidden_act": "silu",
|
| 33 |
+
"hidden_size": 896,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1"
|
| 37 |
+
},
|
| 38 |
+
"initializer_range": 0.02,
|
| 39 |
+
"intermediate_size": 4864,
|
| 40 |
+
"is_decoder": false,
|
| 41 |
+
"is_encoder_decoder": false,
|
| 42 |
+
"label2id": {
|
| 43 |
+
"LABEL_0": 0,
|
| 44 |
+
"LABEL_1": 1
|
| 45 |
+
},
|
| 46 |
+
"length_penalty": 1.0,
|
| 47 |
+
"max_length": 20,
|
| 48 |
+
"max_position_embeddings": 131072,
|
| 49 |
+
"max_window_layers": 24,
|
| 50 |
+
"min_length": 0,
|
| 51 |
+
"model_type": "qwen2",
|
| 52 |
+
"no_repeat_ngram_size": 0,
|
| 53 |
+
"num_attention_heads": 14,
|
| 54 |
+
"num_beam_groups": 1,
|
| 55 |
+
"num_beams": 1,
|
| 56 |
+
"num_hidden_layers": 24,
|
| 57 |
+
"num_key_value_heads": 2,
|
| 58 |
+
"num_return_sequences": 1,
|
| 59 |
+
"output_attentions": false,
|
| 60 |
+
"output_hidden_states": false,
|
| 61 |
+
"output_scores": false,
|
| 62 |
+
"pad_token_id": null,
|
| 63 |
+
"prefix": null,
|
| 64 |
+
"problem_type": null,
|
| 65 |
+
"pruned_heads": {},
|
| 66 |
+
"remove_invalid_values": false,
|
| 67 |
+
"repetition_penalty": 1.0,
|
| 68 |
+
"return_dict": true,
|
| 69 |
+
"return_dict_in_generate": false,
|
| 70 |
+
"rms_norm_eps": 1e-06,
|
| 71 |
+
"rope_theta": 1000000.0,
|
| 72 |
+
"sep_token_id": null,
|
| 73 |
+
"sliding_window": null,
|
| 74 |
+
"suppress_tokens": null,
|
| 75 |
+
"task_specific_params": null,
|
| 76 |
+
"temperature": 1.0,
|
| 77 |
+
"tf_legacy_loss": false,
|
| 78 |
+
"tie_encoder_decoder": false,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_class": null,
|
| 81 |
+
"top_k": 50,
|
| 82 |
+
"top_p": 1.0,
|
| 83 |
+
"torch_dtype": "float32",
|
| 84 |
+
"torchscript": false,
|
| 85 |
+
"typical_p": 1.0,
|
| 86 |
+
"use_bfloat16": false,
|
| 87 |
+
"use_cache": true,
|
| 88 |
+
"use_sliding_window": false,
|
| 89 |
+
"vocab_size": 151648
|
| 90 |
+
},
|
| 91 |
+
"encoder_model_name": "knowledgator/Qwen-encoder-0.5B",
|
| 92 |
+
"extract_text_features": true,
|
| 93 |
+
"focal_loss_alpha": -1,
|
| 94 |
+
"focal_loss_gamma": 2,
|
| 95 |
+
"hidden_size": 896,
|
| 96 |
+
"ignore_index": -100,
|
| 97 |
+
"initializer_range": 0.03,
|
| 98 |
+
"logit_scale_init_value": 2.6592,
|
| 99 |
+
"max_num_classes": 25,
|
| 100 |
+
"model_type": "GLiClass",
|
| 101 |
+
"normalize_features": false,
|
| 102 |
+
"pooling_strategy": "avg",
|
| 103 |
+
"problem_type": "multi_label_classification",
|
| 104 |
+
"projector_hidden_act": "gelu",
|
| 105 |
+
"prompt_first": true,
|
| 106 |
+
"scorer_type": "simple",
|
| 107 |
+
"squeeze_layers": false,
|
| 108 |
+
"text_token_index": 151647,
|
| 109 |
+
"torch_dtype": "float32",
|
| 110 |
+
"transformers_version": "4.44.1",
|
| 111 |
+
"use_lstm": false,
|
| 112 |
+
"vocab_size": 151648
|
| 113 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9c1036a1758ea07a6062b09a17967bfc3a07fb15a4434df1fba378d6ba3f8c5
|
| 3 |
+
size 1987995684
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"mask_token": {
|
| 14 |
+
"content": "_",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
},
|
| 20 |
+
"pad_token": {
|
| 21 |
+
"content": "<|endoftext|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false
|
| 26 |
+
}
|
| 27 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"62": {
|
| 5 |
+
"content": "_",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151643": {
|
| 13 |
+
"content": "<|endoftext|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151644": {
|
| 21 |
+
"content": "<|im_start|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"151645": {
|
| 29 |
+
"content": "<|im_end|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"151646": {
|
| 37 |
+
"content": "<<LABEL>>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"151647": {
|
| 45 |
+
"content": "<<SEP>>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
}
|
| 52 |
+
},
|
| 53 |
+
"additional_special_tokens": [
|
| 54 |
+
"<|im_start|>",
|
| 55 |
+
"<|im_end|>"
|
| 56 |
+
],
|
| 57 |
+
"bos_token": null,
|
| 58 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 59 |
+
"clean_up_tokenization_spaces": false,
|
| 60 |
+
"eos_token": "<|endoftext|>",
|
| 61 |
+
"errors": "replace",
|
| 62 |
+
"mask_token": "_",
|
| 63 |
+
"model_max_length": 32768,
|
| 64 |
+
"pad_token": "<|endoftext|>",
|
| 65 |
+
"split_special_tokens": false,
|
| 66 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 67 |
+
"unk_token": null
|
| 68 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2145 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.7980221756068326,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 30000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005993407252022775,
|
| 13 |
+
"grad_norm": 2.2955188751220703,
|
| 14 |
+
"learning_rate": 2.996703626011388e-07,
|
| 15 |
+
"loss": 0.2171,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01198681450404555,
|
| 20 |
+
"grad_norm": 0.9237516522407532,
|
| 21 |
+
"learning_rate": 5.993407252022776e-07,
|
| 22 |
+
"loss": 0.139,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.017980221756068324,
|
| 27 |
+
"grad_norm": 0.4022263288497925,
|
| 28 |
+
"learning_rate": 8.990110878034162e-07,
|
| 29 |
+
"loss": 0.1111,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0239736290080911,
|
| 34 |
+
"grad_norm": 0.28793156147003174,
|
| 35 |
+
"learning_rate": 1.1986814504045551e-06,
|
| 36 |
+
"loss": 0.1079,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.029967036260113874,
|
| 41 |
+
"grad_norm": 0.479831337928772,
|
| 42 |
+
"learning_rate": 1.4983518130056939e-06,
|
| 43 |
+
"loss": 0.1078,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03596044351213665,
|
| 48 |
+
"grad_norm": 0.2760140895843506,
|
| 49 |
+
"learning_rate": 1.7980221756068325e-06,
|
| 50 |
+
"loss": 0.1052,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.041953850764159424,
|
| 55 |
+
"grad_norm": 0.44269922375679016,
|
| 56 |
+
"learning_rate": 2.0976925382079712e-06,
|
| 57 |
+
"loss": 0.1052,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0479472580161822,
|
| 62 |
+
"grad_norm": 0.25097307562828064,
|
| 63 |
+
"learning_rate": 2.3973629008091102e-06,
|
| 64 |
+
"loss": 0.1054,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05394066526820498,
|
| 69 |
+
"grad_norm": 0.3348759412765503,
|
| 70 |
+
"learning_rate": 2.6970332634102492e-06,
|
| 71 |
+
"loss": 0.1042,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.05993407252022775,
|
| 76 |
+
"grad_norm": 0.11476008594036102,
|
| 77 |
+
"learning_rate": 2.9967036260113878e-06,
|
| 78 |
+
"loss": 0.104,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06592747977225052,
|
| 83 |
+
"grad_norm": 0.1612076610326767,
|
| 84 |
+
"learning_rate": 3.2963739886125264e-06,
|
| 85 |
+
"loss": 0.1041,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0719208870242733,
|
| 90 |
+
"grad_norm": 0.3466070592403412,
|
| 91 |
+
"learning_rate": 3.596044351213665e-06,
|
| 92 |
+
"loss": 0.1034,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.07791429427629608,
|
| 97 |
+
"grad_norm": 0.41668832302093506,
|
| 98 |
+
"learning_rate": 3.8957147138148035e-06,
|
| 99 |
+
"loss": 0.1023,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08390770152831885,
|
| 104 |
+
"grad_norm": 0.3885754346847534,
|
| 105 |
+
"learning_rate": 4.1953850764159425e-06,
|
| 106 |
+
"loss": 0.1038,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.08990110878034162,
|
| 111 |
+
"grad_norm": 0.3091404438018799,
|
| 112 |
+
"learning_rate": 4.4950554390170815e-06,
|
| 113 |
+
"loss": 0.1022,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0958945160323644,
|
| 118 |
+
"grad_norm": 0.24296438694000244,
|
| 119 |
+
"learning_rate": 4.7947258016182205e-06,
|
| 120 |
+
"loss": 0.1014,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.10188792328438717,
|
| 125 |
+
"grad_norm": 0.2710008919239044,
|
| 126 |
+
"learning_rate": 5.0943961642193594e-06,
|
| 127 |
+
"loss": 0.1014,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.10788133053640996,
|
| 132 |
+
"grad_norm": 0.10421634465456009,
|
| 133 |
+
"learning_rate": 5.3940665268204984e-06,
|
| 134 |
+
"loss": 0.1006,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.11387473778843273,
|
| 139 |
+
"grad_norm": 0.35064780712127686,
|
| 140 |
+
"learning_rate": 5.6937368894216374e-06,
|
| 141 |
+
"loss": 0.1006,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1198681450404555,
|
| 146 |
+
"grad_norm": 0.3558998703956604,
|
| 147 |
+
"learning_rate": 5.9934072520227756e-06,
|
| 148 |
+
"loss": 0.101,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.12586155229247828,
|
| 153 |
+
"grad_norm": 0.13521741330623627,
|
| 154 |
+
"learning_rate": 6.2930776146239146e-06,
|
| 155 |
+
"loss": 0.101,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.13185495954450105,
|
| 160 |
+
"grad_norm": 0.23776298761367798,
|
| 161 |
+
"learning_rate": 6.592747977225053e-06,
|
| 162 |
+
"loss": 0.1013,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.13784836679652382,
|
| 167 |
+
"grad_norm": 0.12444166839122772,
|
| 168 |
+
"learning_rate": 6.892418339826191e-06,
|
| 169 |
+
"loss": 0.1008,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1438417740485466,
|
| 174 |
+
"grad_norm": 0.13655610382556915,
|
| 175 |
+
"learning_rate": 7.19208870242733e-06,
|
| 176 |
+
"loss": 0.1019,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.1498351813005694,
|
| 181 |
+
"grad_norm": 0.27996689081192017,
|
| 182 |
+
"learning_rate": 7.491759065028469e-06,
|
| 183 |
+
"loss": 0.1014,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.15582858855259216,
|
| 188 |
+
"grad_norm": 0.30384400486946106,
|
| 189 |
+
"learning_rate": 7.791429427629607e-06,
|
| 190 |
+
"loss": 0.1005,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.16182199580461493,
|
| 195 |
+
"grad_norm": 0.06711748242378235,
|
| 196 |
+
"learning_rate": 8.091099790230747e-06,
|
| 197 |
+
"loss": 0.1001,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1678154030566377,
|
| 202 |
+
"grad_norm": 0.20208857953548431,
|
| 203 |
+
"learning_rate": 8.390770152831885e-06,
|
| 204 |
+
"loss": 0.1011,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.17380881030866047,
|
| 209 |
+
"grad_norm": 0.5147479772567749,
|
| 210 |
+
"learning_rate": 8.690440515433025e-06,
|
| 211 |
+
"loss": 0.1012,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.17980221756068324,
|
| 216 |
+
"grad_norm": 0.2111215740442276,
|
| 217 |
+
"learning_rate": 8.990110878034163e-06,
|
| 218 |
+
"loss": 0.1002,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.18579562481270603,
|
| 223 |
+
"grad_norm": 0.3195187449455261,
|
| 224 |
+
"learning_rate": 9.289781240635301e-06,
|
| 225 |
+
"loss": 0.1005,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1917890320647288,
|
| 230 |
+
"grad_norm": 0.1801885962486267,
|
| 231 |
+
"learning_rate": 9.589451603236441e-06,
|
| 232 |
+
"loss": 0.0999,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.19778243931675157,
|
| 237 |
+
"grad_norm": 0.07530693709850311,
|
| 238 |
+
"learning_rate": 9.889121965837579e-06,
|
| 239 |
+
"loss": 0.1019,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.20377584656877434,
|
| 244 |
+
"grad_norm": 0.13042743504047394,
|
| 245 |
+
"learning_rate": 9.97902307461792e-06,
|
| 246 |
+
"loss": 0.1003,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.2097692538207971,
|
| 251 |
+
"grad_norm": 0.26616647839546204,
|
| 252 |
+
"learning_rate": 9.94572636766224e-06,
|
| 253 |
+
"loss": 0.1007,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2157626610728199,
|
| 258 |
+
"grad_norm": 0.35827577114105225,
|
| 259 |
+
"learning_rate": 9.912429660706557e-06,
|
| 260 |
+
"loss": 0.1007,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.22175606832484268,
|
| 265 |
+
"grad_norm": 0.30124035477638245,
|
| 266 |
+
"learning_rate": 9.879132953750875e-06,
|
| 267 |
+
"loss": 0.1,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.22774947557686545,
|
| 272 |
+
"grad_norm": 0.34060993790626526,
|
| 273 |
+
"learning_rate": 9.845836246795192e-06,
|
| 274 |
+
"loss": 0.1013,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.23374288282888822,
|
| 279 |
+
"grad_norm": 0.14318062365055084,
|
| 280 |
+
"learning_rate": 9.812539539839511e-06,
|
| 281 |
+
"loss": 0.1008,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.239736290080911,
|
| 286 |
+
"grad_norm": 0.310312420129776,
|
| 287 |
+
"learning_rate": 9.779242832883829e-06,
|
| 288 |
+
"loss": 0.0999,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.24572969733293376,
|
| 293 |
+
"grad_norm": 0.1943434178829193,
|
| 294 |
+
"learning_rate": 9.745946125928146e-06,
|
| 295 |
+
"loss": 0.0999,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.25172310458495656,
|
| 300 |
+
"grad_norm": 0.22400355339050293,
|
| 301 |
+
"learning_rate": 9.712649418972464e-06,
|
| 302 |
+
"loss": 0.1016,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2577165118369793,
|
| 307 |
+
"grad_norm": 0.13477976620197296,
|
| 308 |
+
"learning_rate": 9.679352712016781e-06,
|
| 309 |
+
"loss": 0.101,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2637099190890021,
|
| 314 |
+
"grad_norm": 0.15206047892570496,
|
| 315 |
+
"learning_rate": 9.6460560050611e-06,
|
| 316 |
+
"loss": 0.1009,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2697033263410249,
|
| 321 |
+
"grad_norm": 0.060487061738967896,
|
| 322 |
+
"learning_rate": 9.61275929810542e-06,
|
| 323 |
+
"loss": 0.0996,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.27569673359304764,
|
| 328 |
+
"grad_norm": 0.16848477721214294,
|
| 329 |
+
"learning_rate": 9.579462591149737e-06,
|
| 330 |
+
"loss": 0.0998,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.28169014084507044,
|
| 335 |
+
"grad_norm": 0.10402075946331024,
|
| 336 |
+
"learning_rate": 9.546165884194054e-06,
|
| 337 |
+
"loss": 0.1012,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2876835480970932,
|
| 342 |
+
"grad_norm": 0.12113852798938751,
|
| 343 |
+
"learning_rate": 9.512869177238371e-06,
|
| 344 |
+
"loss": 0.1008,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.293676955349116,
|
| 349 |
+
"grad_norm": 0.23850949108600616,
|
| 350 |
+
"learning_rate": 9.479572470282689e-06,
|
| 351 |
+
"loss": 0.1005,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2996703626011388,
|
| 356 |
+
"grad_norm": 0.3209283649921417,
|
| 357 |
+
"learning_rate": 9.446275763327008e-06,
|
| 358 |
+
"loss": 0.1005,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.3056637698531615,
|
| 363 |
+
"grad_norm": 0.6144764423370361,
|
| 364 |
+
"learning_rate": 9.412979056371325e-06,
|
| 365 |
+
"loss": 0.1,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3116571771051843,
|
| 370 |
+
"grad_norm": 0.04221544787287712,
|
| 371 |
+
"learning_rate": 9.379682349415643e-06,
|
| 372 |
+
"loss": 0.0997,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.31765058435720706,
|
| 377 |
+
"grad_norm": 0.07346853613853455,
|
| 378 |
+
"learning_rate": 9.346385642459962e-06,
|
| 379 |
+
"loss": 0.1011,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.32364399160922985,
|
| 384 |
+
"grad_norm": 0.3777712285518646,
|
| 385 |
+
"learning_rate": 9.31308893550428e-06,
|
| 386 |
+
"loss": 0.1008,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3296373988612526,
|
| 391 |
+
"grad_norm": 0.07576871663331985,
|
| 392 |
+
"learning_rate": 9.279792228548597e-06,
|
| 393 |
+
"loss": 0.1001,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3356308061132754,
|
| 398 |
+
"grad_norm": 0.194047212600708,
|
| 399 |
+
"learning_rate": 9.246495521592916e-06,
|
| 400 |
+
"loss": 0.1003,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3416242133652982,
|
| 405 |
+
"grad_norm": 0.09262324869632721,
|
| 406 |
+
"learning_rate": 9.213198814637233e-06,
|
| 407 |
+
"loss": 0.0995,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.34761762061732093,
|
| 412 |
+
"grad_norm": 0.5301212668418884,
|
| 413 |
+
"learning_rate": 9.17990210768155e-06,
|
| 414 |
+
"loss": 0.0999,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.35361102786934373,
|
| 419 |
+
"grad_norm": 0.04346400871872902,
|
| 420 |
+
"learning_rate": 9.146605400725868e-06,
|
| 421 |
+
"loss": 0.0995,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3596044351213665,
|
| 426 |
+
"grad_norm": 0.16581957042217255,
|
| 427 |
+
"learning_rate": 9.113308693770187e-06,
|
| 428 |
+
"loss": 0.0995,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.36559784237338927,
|
| 433 |
+
"grad_norm": 0.37740394473075867,
|
| 434 |
+
"learning_rate": 9.080011986814505e-06,
|
| 435 |
+
"loss": 0.1005,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.37159124962541207,
|
| 440 |
+
"grad_norm": 0.08017630875110626,
|
| 441 |
+
"learning_rate": 9.046715279858824e-06,
|
| 442 |
+
"loss": 0.1,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.3775846568774348,
|
| 447 |
+
"grad_norm": 0.1393900364637375,
|
| 448 |
+
"learning_rate": 9.013418572903141e-06,
|
| 449 |
+
"loss": 0.1007,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3835780641294576,
|
| 454 |
+
"grad_norm": 0.11385737359523773,
|
| 455 |
+
"learning_rate": 8.980121865947459e-06,
|
| 456 |
+
"loss": 0.1001,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.38957147138148035,
|
| 461 |
+
"grad_norm": 0.1457832008600235,
|
| 462 |
+
"learning_rate": 8.946825158991776e-06,
|
| 463 |
+
"loss": 0.1,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.39556487863350315,
|
| 468 |
+
"grad_norm": 0.03458058834075928,
|
| 469 |
+
"learning_rate": 8.913528452036094e-06,
|
| 470 |
+
"loss": 0.0999,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.40155828588552595,
|
| 475 |
+
"grad_norm": 0.4892089366912842,
|
| 476 |
+
"learning_rate": 8.880231745080413e-06,
|
| 477 |
+
"loss": 0.1,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.4075516931375487,
|
| 482 |
+
"grad_norm": 0.11881079524755478,
|
| 483 |
+
"learning_rate": 8.84693503812473e-06,
|
| 484 |
+
"loss": 0.1009,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.4135451003895715,
|
| 489 |
+
"grad_norm": 0.29624250531196594,
|
| 490 |
+
"learning_rate": 8.813638331169047e-06,
|
| 491 |
+
"loss": 0.1007,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.4195385076415942,
|
| 496 |
+
"grad_norm": 0.18601752817630768,
|
| 497 |
+
"learning_rate": 8.780341624213367e-06,
|
| 498 |
+
"loss": 0.1002,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.425531914893617,
|
| 503 |
+
"grad_norm": 0.017990827560424805,
|
| 504 |
+
"learning_rate": 8.747044917257684e-06,
|
| 505 |
+
"loss": 0.0998,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4315253221456398,
|
| 510 |
+
"grad_norm": 0.08424385637044907,
|
| 511 |
+
"learning_rate": 8.713748210302001e-06,
|
| 512 |
+
"loss": 0.1005,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.43751872939766256,
|
| 517 |
+
"grad_norm": 0.40114349126815796,
|
| 518 |
+
"learning_rate": 8.68045150334632e-06,
|
| 519 |
+
"loss": 0.0994,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.44351213664968536,
|
| 524 |
+
"grad_norm": 0.10041668266057968,
|
| 525 |
+
"learning_rate": 8.647154796390638e-06,
|
| 526 |
+
"loss": 0.0999,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.4495055439017081,
|
| 531 |
+
"grad_norm": 0.4256216585636139,
|
| 532 |
+
"learning_rate": 8.613858089434955e-06,
|
| 533 |
+
"loss": 0.101,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.4554989511537309,
|
| 538 |
+
"grad_norm": 0.2503170371055603,
|
| 539 |
+
"learning_rate": 8.580561382479273e-06,
|
| 540 |
+
"loss": 0.0996,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.46149235840575364,
|
| 545 |
+
"grad_norm": 0.09117023646831512,
|
| 546 |
+
"learning_rate": 8.547264675523592e-06,
|
| 547 |
+
"loss": 0.0987,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.46748576565777644,
|
| 552 |
+
"grad_norm": 0.12042222917079926,
|
| 553 |
+
"learning_rate": 8.51396796856791e-06,
|
| 554 |
+
"loss": 0.1002,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.47347917290979924,
|
| 559 |
+
"grad_norm": 0.13141508400440216,
|
| 560 |
+
"learning_rate": 8.480671261612228e-06,
|
| 561 |
+
"loss": 0.101,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.479472580161822,
|
| 566 |
+
"grad_norm": 0.20164711773395538,
|
| 567 |
+
"learning_rate": 8.447374554656546e-06,
|
| 568 |
+
"loss": 0.101,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4854659874138448,
|
| 573 |
+
"grad_norm": 0.09953387826681137,
|
| 574 |
+
"learning_rate": 8.414077847700863e-06,
|
| 575 |
+
"loss": 0.0998,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4914593946658675,
|
| 580 |
+
"grad_norm": 0.03517329692840576,
|
| 581 |
+
"learning_rate": 8.38078114074518e-06,
|
| 582 |
+
"loss": 0.1008,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4974528019178903,
|
| 587 |
+
"grad_norm": 0.01899944245815277,
|
| 588 |
+
"learning_rate": 8.347484433789498e-06,
|
| 589 |
+
"loss": 0.0995,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5034462091699131,
|
| 594 |
+
"grad_norm": 0.20903293788433075,
|
| 595 |
+
"learning_rate": 8.314187726833817e-06,
|
| 596 |
+
"loss": 0.0996,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5094396164219359,
|
| 601 |
+
"grad_norm": 0.28464946150779724,
|
| 602 |
+
"learning_rate": 8.280891019878135e-06,
|
| 603 |
+
"loss": 0.1007,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5154330236739586,
|
| 608 |
+
"grad_norm": 0.2716582417488098,
|
| 609 |
+
"learning_rate": 8.247594312922452e-06,
|
| 610 |
+
"loss": 0.1001,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5214264309259814,
|
| 615 |
+
"grad_norm": 0.07964222133159637,
|
| 616 |
+
"learning_rate": 8.21429760596677e-06,
|
| 617 |
+
"loss": 0.1,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5274198381780042,
|
| 622 |
+
"grad_norm": 0.0397951677441597,
|
| 623 |
+
"learning_rate": 8.181000899011089e-06,
|
| 624 |
+
"loss": 0.1003,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.533413245430027,
|
| 629 |
+
"grad_norm": 0.31348341703414917,
|
| 630 |
+
"learning_rate": 8.147704192055406e-06,
|
| 631 |
+
"loss": 0.1001,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5394066526820498,
|
| 636 |
+
"grad_norm": 0.0627601370215416,
|
| 637 |
+
"learning_rate": 8.114407485099725e-06,
|
| 638 |
+
"loss": 0.1003,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.5454000599340725,
|
| 643 |
+
"grad_norm": 0.2744080424308777,
|
| 644 |
+
"learning_rate": 8.081110778144043e-06,
|
| 645 |
+
"loss": 0.1002,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.5513934671860953,
|
| 650 |
+
"grad_norm": 0.14465458691120148,
|
| 651 |
+
"learning_rate": 8.04781407118836e-06,
|
| 652 |
+
"loss": 0.0998,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5573868744381181,
|
| 657 |
+
"grad_norm": 0.21187633275985718,
|
| 658 |
+
"learning_rate": 8.014517364232677e-06,
|
| 659 |
+
"loss": 0.0999,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5633802816901409,
|
| 664 |
+
"grad_norm": 0.06333980709314346,
|
| 665 |
+
"learning_rate": 7.981220657276996e-06,
|
| 666 |
+
"loss": 0.0998,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5693736889421637,
|
| 671 |
+
"grad_norm": 0.3627989888191223,
|
| 672 |
+
"learning_rate": 7.947923950321314e-06,
|
| 673 |
+
"loss": 0.0996,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5753670961941864,
|
| 678 |
+
"grad_norm": 0.11346852034330368,
|
| 679 |
+
"learning_rate": 7.914627243365631e-06,
|
| 680 |
+
"loss": 0.1008,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5813605034462092,
|
| 685 |
+
"grad_norm": 0.1832839399576187,
|
| 686 |
+
"learning_rate": 7.88133053640995e-06,
|
| 687 |
+
"loss": 0.1,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.587353910698232,
|
| 692 |
+
"grad_norm": 0.034973230212926865,
|
| 693 |
+
"learning_rate": 7.848033829454268e-06,
|
| 694 |
+
"loss": 0.1,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5933473179502547,
|
| 699 |
+
"grad_norm": 0.11853715777397156,
|
| 700 |
+
"learning_rate": 7.814737122498585e-06,
|
| 701 |
+
"loss": 0.0991,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5993407252022775,
|
| 706 |
+
"grad_norm": 0.13182196021080017,
|
| 707 |
+
"learning_rate": 7.781440415542903e-06,
|
| 708 |
+
"loss": 0.0993,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.6053341324543002,
|
| 713 |
+
"grad_norm": 0.09213405847549438,
|
| 714 |
+
"learning_rate": 7.748143708587222e-06,
|
| 715 |
+
"loss": 0.0997,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.611327539706323,
|
| 720 |
+
"grad_norm": 0.06769004464149475,
|
| 721 |
+
"learning_rate": 7.71484700163154e-06,
|
| 722 |
+
"loss": 0.1008,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6173209469583458,
|
| 727 |
+
"grad_norm": 0.020326493307948112,
|
| 728 |
+
"learning_rate": 7.681550294675857e-06,
|
| 729 |
+
"loss": 0.0993,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.6233143542103686,
|
| 734 |
+
"grad_norm": 0.04122787341475487,
|
| 735 |
+
"learning_rate": 7.648253587720174e-06,
|
| 736 |
+
"loss": 0.0997,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.6293077614623914,
|
| 741 |
+
"grad_norm": 0.1741415411233902,
|
| 742 |
+
"learning_rate": 7.614956880764492e-06,
|
| 743 |
+
"loss": 0.0998,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.6353011687144141,
|
| 748 |
+
"grad_norm": 0.028361910954117775,
|
| 749 |
+
"learning_rate": 7.5816601738088115e-06,
|
| 750 |
+
"loss": 0.0992,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.6412945759664369,
|
| 755 |
+
"grad_norm": 0.10101690888404846,
|
| 756 |
+
"learning_rate": 7.548363466853129e-06,
|
| 757 |
+
"loss": 0.0987,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.6472879832184597,
|
| 762 |
+
"grad_norm": 0.1292373090982437,
|
| 763 |
+
"learning_rate": 7.515066759897447e-06,
|
| 764 |
+
"loss": 0.101,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.6532813904704825,
|
| 769 |
+
"grad_norm": 0.25996342301368713,
|
| 770 |
+
"learning_rate": 7.4817700529417646e-06,
|
| 771 |
+
"loss": 0.0998,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.6592747977225052,
|
| 776 |
+
"grad_norm": 0.08856533467769623,
|
| 777 |
+
"learning_rate": 7.448473345986083e-06,
|
| 778 |
+
"loss": 0.0993,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.665268204974528,
|
| 783 |
+
"grad_norm": 0.11309339106082916,
|
| 784 |
+
"learning_rate": 7.4151766390304e-06,
|
| 785 |
+
"loss": 0.099,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.6712616122265508,
|
| 790 |
+
"grad_norm": 0.28294652700424194,
|
| 791 |
+
"learning_rate": 7.381879932074718e-06,
|
| 792 |
+
"loss": 0.1009,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6772550194785736,
|
| 797 |
+
"grad_norm": 0.18077927827835083,
|
| 798 |
+
"learning_rate": 7.348583225119036e-06,
|
| 799 |
+
"loss": 0.1004,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6832484267305964,
|
| 804 |
+
"grad_norm": 0.03750494495034218,
|
| 805 |
+
"learning_rate": 7.315286518163355e-06,
|
| 806 |
+
"loss": 0.0996,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6892418339826191,
|
| 811 |
+
"grad_norm": 0.11316617578268051,
|
| 812 |
+
"learning_rate": 7.2819898112076725e-06,
|
| 813 |
+
"loss": 0.1003,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6952352412346419,
|
| 818 |
+
"grad_norm": 0.06800729036331177,
|
| 819 |
+
"learning_rate": 7.248693104251991e-06,
|
| 820 |
+
"loss": 0.0998,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.7012286484866647,
|
| 825 |
+
"grad_norm": 0.024295495823025703,
|
| 826 |
+
"learning_rate": 7.215396397296308e-06,
|
| 827 |
+
"loss": 0.1006,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.7072220557386875,
|
| 832 |
+
"grad_norm": 0.03560725226998329,
|
| 833 |
+
"learning_rate": 7.1820996903406256e-06,
|
| 834 |
+
"loss": 0.0996,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.7132154629907103,
|
| 839 |
+
"grad_norm": 0.08191557973623276,
|
| 840 |
+
"learning_rate": 7.148802983384944e-06,
|
| 841 |
+
"loss": 0.0998,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.719208870242733,
|
| 846 |
+
"grad_norm": 0.1713048815727234,
|
| 847 |
+
"learning_rate": 7.115506276429261e-06,
|
| 848 |
+
"loss": 0.1,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.7252022774947557,
|
| 853 |
+
"grad_norm": 0.04900109022855759,
|
| 854 |
+
"learning_rate": 7.0822095694735795e-06,
|
| 855 |
+
"loss": 0.0996,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.7311956847467785,
|
| 860 |
+
"grad_norm": 0.04907857999205589,
|
| 861 |
+
"learning_rate": 7.048912862517897e-06,
|
| 862 |
+
"loss": 0.0999,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.7371890919988013,
|
| 867 |
+
"grad_norm": 0.07542908936738968,
|
| 868 |
+
"learning_rate": 7.015616155562216e-06,
|
| 869 |
+
"loss": 0.0988,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.7431824992508241,
|
| 874 |
+
"grad_norm": 0.0938209518790245,
|
| 875 |
+
"learning_rate": 6.9823194486065335e-06,
|
| 876 |
+
"loss": 0.0992,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.7491759065028468,
|
| 881 |
+
"grad_norm": 0.02947254106402397,
|
| 882 |
+
"learning_rate": 6.949022741650852e-06,
|
| 883 |
+
"loss": 0.0983,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.7551693137548696,
|
| 888 |
+
"grad_norm": 0.4940189719200134,
|
| 889 |
+
"learning_rate": 6.915726034695169e-06,
|
| 890 |
+
"loss": 0.0992,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.7611627210068924,
|
| 895 |
+
"grad_norm": 2.7082247734069824,
|
| 896 |
+
"learning_rate": 6.8824293277394874e-06,
|
| 897 |
+
"loss": 0.0988,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.7671561282589152,
|
| 902 |
+
"grad_norm": 0.427272766828537,
|
| 903 |
+
"learning_rate": 6.849132620783805e-06,
|
| 904 |
+
"loss": 0.0999,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.773149535510938,
|
| 909 |
+
"grad_norm": 0.11108270287513733,
|
| 910 |
+
"learning_rate": 6.815835913828123e-06,
|
| 911 |
+
"loss": 0.1006,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.7791429427629607,
|
| 916 |
+
"grad_norm": 0.05636779963970184,
|
| 917 |
+
"learning_rate": 6.7825392068724405e-06,
|
| 918 |
+
"loss": 0.0998,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.7851363500149835,
|
| 923 |
+
"grad_norm": 0.2918495237827301,
|
| 924 |
+
"learning_rate": 6.749242499916758e-06,
|
| 925 |
+
"loss": 0.0999,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.7911297572670063,
|
| 930 |
+
"grad_norm": 0.08755291998386383,
|
| 931 |
+
"learning_rate": 6.715945792961077e-06,
|
| 932 |
+
"loss": 0.1006,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7971231645190291,
|
| 937 |
+
"grad_norm": 0.11561553925275803,
|
| 938 |
+
"learning_rate": 6.682649086005395e-06,
|
| 939 |
+
"loss": 0.0994,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.8031165717710519,
|
| 944 |
+
"grad_norm": 0.03734762221574783,
|
| 945 |
+
"learning_rate": 6.649352379049713e-06,
|
| 946 |
+
"loss": 0.0989,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.8091099790230746,
|
| 951 |
+
"grad_norm": 0.07186517119407654,
|
| 952 |
+
"learning_rate": 6.61605567209403e-06,
|
| 953 |
+
"loss": 0.099,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.8151033862750974,
|
| 958 |
+
"grad_norm": 0.29982343316078186,
|
| 959 |
+
"learning_rate": 6.5827589651383484e-06,
|
| 960 |
+
"loss": 0.1,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.8210967935271202,
|
| 965 |
+
"grad_norm": 0.09515996277332306,
|
| 966 |
+
"learning_rate": 6.549462258182666e-06,
|
| 967 |
+
"loss": 0.0992,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.827090200779143,
|
| 972 |
+
"grad_norm": 0.07374672591686249,
|
| 973 |
+
"learning_rate": 6.516165551226984e-06,
|
| 974 |
+
"loss": 0.0992,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.8330836080311658,
|
| 979 |
+
"grad_norm": 0.20344111323356628,
|
| 980 |
+
"learning_rate": 6.4828688442713015e-06,
|
| 981 |
+
"loss": 0.1004,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.8390770152831885,
|
| 986 |
+
"grad_norm": 0.43719688057899475,
|
| 987 |
+
"learning_rate": 6.44957213731562e-06,
|
| 988 |
+
"loss": 0.0994,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.8450704225352113,
|
| 993 |
+
"grad_norm": 0.06093699112534523,
|
| 994 |
+
"learning_rate": 6.416275430359938e-06,
|
| 995 |
+
"loss": 0.0997,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.851063829787234,
|
| 1000 |
+
"grad_norm": 0.25451895594596863,
|
| 1001 |
+
"learning_rate": 6.382978723404256e-06,
|
| 1002 |
+
"loss": 0.0991,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.8570572370392568,
|
| 1007 |
+
"grad_norm": 0.09084396809339523,
|
| 1008 |
+
"learning_rate": 6.349682016448574e-06,
|
| 1009 |
+
"loss": 0.1,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.8630506442912796,
|
| 1014 |
+
"grad_norm": 0.1228547915816307,
|
| 1015 |
+
"learning_rate": 6.316385309492892e-06,
|
| 1016 |
+
"loss": 0.0997,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.8690440515433023,
|
| 1021 |
+
"grad_norm": 0.27622973918914795,
|
| 1022 |
+
"learning_rate": 6.2830886025372095e-06,
|
| 1023 |
+
"loss": 0.0989,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.8750374587953251,
|
| 1028 |
+
"grad_norm": 0.11149102449417114,
|
| 1029 |
+
"learning_rate": 6.249791895581528e-06,
|
| 1030 |
+
"loss": 0.1002,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.8810308660473479,
|
| 1035 |
+
"grad_norm": 0.10650242865085602,
|
| 1036 |
+
"learning_rate": 6.216495188625845e-06,
|
| 1037 |
+
"loss": 0.0994,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.8870242732993707,
|
| 1042 |
+
"grad_norm": 0.46797627210617065,
|
| 1043 |
+
"learning_rate": 6.1831984816701626e-06,
|
| 1044 |
+
"loss": 0.1005,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.8930176805513935,
|
| 1049 |
+
"grad_norm": 0.46854981780052185,
|
| 1050 |
+
"learning_rate": 6.149901774714481e-06,
|
| 1051 |
+
"loss": 0.0988,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.8990110878034162,
|
| 1056 |
+
"grad_norm": 0.12268362194299698,
|
| 1057 |
+
"learning_rate": 6.1166050677588e-06,
|
| 1058 |
+
"loss": 0.0988,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.905004495055439,
|
| 1063 |
+
"grad_norm": 0.13641001284122467,
|
| 1064 |
+
"learning_rate": 6.083308360803117e-06,
|
| 1065 |
+
"loss": 0.099,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.9109979023074618,
|
| 1070 |
+
"grad_norm": 0.014134293422102928,
|
| 1071 |
+
"learning_rate": 6.050011653847435e-06,
|
| 1072 |
+
"loss": 0.0983,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.9169913095594846,
|
| 1077 |
+
"grad_norm": 0.1600305587053299,
|
| 1078 |
+
"learning_rate": 6.016714946891753e-06,
|
| 1079 |
+
"loss": 0.0984,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.9229847168115073,
|
| 1084 |
+
"grad_norm": 0.04153376817703247,
|
| 1085 |
+
"learning_rate": 5.9834182399360705e-06,
|
| 1086 |
+
"loss": 0.0998,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.9289781240635301,
|
| 1091 |
+
"grad_norm": 0.364000529050827,
|
| 1092 |
+
"learning_rate": 5.950121532980389e-06,
|
| 1093 |
+
"loss": 0.099,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.9349715313155529,
|
| 1098 |
+
"grad_norm": 0.10466318577528,
|
| 1099 |
+
"learning_rate": 5.916824826024706e-06,
|
| 1100 |
+
"loss": 0.0997,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.9409649385675757,
|
| 1105 |
+
"grad_norm": 0.2283148467540741,
|
| 1106 |
+
"learning_rate": 5.883528119069024e-06,
|
| 1107 |
+
"loss": 0.0996,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.9469583458195985,
|
| 1112 |
+
"grad_norm": 0.07752685993909836,
|
| 1113 |
+
"learning_rate": 5.850231412113343e-06,
|
| 1114 |
+
"loss": 0.1004,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.9529517530716212,
|
| 1119 |
+
"grad_norm": 0.0329587459564209,
|
| 1120 |
+
"learning_rate": 5.816934705157661e-06,
|
| 1121 |
+
"loss": 0.0991,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.958945160323644,
|
| 1126 |
+
"grad_norm": 0.1508065164089203,
|
| 1127 |
+
"learning_rate": 5.783637998201978e-06,
|
| 1128 |
+
"loss": 0.1006,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.9649385675756668,
|
| 1133 |
+
"grad_norm": 0.029579168185591698,
|
| 1134 |
+
"learning_rate": 5.750341291246297e-06,
|
| 1135 |
+
"loss": 0.0991,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.9709319748276896,
|
| 1140 |
+
"grad_norm": 0.029158690944314003,
|
| 1141 |
+
"learning_rate": 5.717044584290614e-06,
|
| 1142 |
+
"loss": 0.0989,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.9769253820797124,
|
| 1147 |
+
"grad_norm": 0.08389827609062195,
|
| 1148 |
+
"learning_rate": 5.683747877334932e-06,
|
| 1149 |
+
"loss": 0.0997,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.982918789331735,
|
| 1154 |
+
"grad_norm": 0.13760380446910858,
|
| 1155 |
+
"learning_rate": 5.65045117037925e-06,
|
| 1156 |
+
"loss": 0.099,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.9889121965837578,
|
| 1161 |
+
"grad_norm": 0.06386566907167435,
|
| 1162 |
+
"learning_rate": 5.617154463423567e-06,
|
| 1163 |
+
"loss": 0.0995,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9949056038357806,
|
| 1168 |
+
"grad_norm": 0.03102271445095539,
|
| 1169 |
+
"learning_rate": 5.583857756467885e-06,
|
| 1170 |
+
"loss": 0.0999,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.0,
|
| 1175 |
+
"eval_accuracy": 0.6769802089043807,
|
| 1176 |
+
"eval_f1": 0.6622248214359344,
|
| 1177 |
+
"eval_loss": 0.0996900200843811,
|
| 1178 |
+
"eval_precision": 0.6552669320986634,
|
| 1179 |
+
"eval_recall": 0.6769802089043807,
|
| 1180 |
+
"eval_runtime": 539.1847,
|
| 1181 |
+
"eval_samples_per_second": 27.506,
|
| 1182 |
+
"eval_steps_per_second": 3.439,
|
| 1183 |
+
"step": 16685
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 1.0008990110878033,
|
| 1187 |
+
"grad_norm": 0.4820581376552582,
|
| 1188 |
+
"learning_rate": 5.5505610495122045e-06,
|
| 1189 |
+
"loss": 0.0985,
|
| 1190 |
+
"step": 16700
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 1.0068924183398262,
|
| 1194 |
+
"grad_norm": 0.35515159368515015,
|
| 1195 |
+
"learning_rate": 5.517264342556522e-06,
|
| 1196 |
+
"loss": 0.0984,
|
| 1197 |
+
"step": 16800
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 1.012885825591849,
|
| 1201 |
+
"grad_norm": 0.043349362909793854,
|
| 1202 |
+
"learning_rate": 5.483967635600839e-06,
|
| 1203 |
+
"loss": 0.0986,
|
| 1204 |
+
"step": 16900
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 1.0188792328438718,
|
| 1208 |
+
"grad_norm": 0.03270228952169418,
|
| 1209 |
+
"learning_rate": 5.450670928645158e-06,
|
| 1210 |
+
"loss": 0.099,
|
| 1211 |
+
"step": 17000
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 1.0248726400958945,
|
| 1215 |
+
"grad_norm": 0.027137137949466705,
|
| 1216 |
+
"learning_rate": 5.417374221689475e-06,
|
| 1217 |
+
"loss": 0.099,
|
| 1218 |
+
"step": 17100
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 1.0308660473479172,
|
| 1222 |
+
"grad_norm": 0.016054777428507805,
|
| 1223 |
+
"learning_rate": 5.384077514733793e-06,
|
| 1224 |
+
"loss": 0.0985,
|
| 1225 |
+
"step": 17200
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 1.0368594545999401,
|
| 1229 |
+
"grad_norm": 0.14413999021053314,
|
| 1230 |
+
"learning_rate": 5.350780807778111e-06,
|
| 1231 |
+
"loss": 0.0988,
|
| 1232 |
+
"step": 17300
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.0428528618519628,
|
| 1236 |
+
"grad_norm": 0.12687167525291443,
|
| 1237 |
+
"learning_rate": 5.317484100822429e-06,
|
| 1238 |
+
"loss": 0.0983,
|
| 1239 |
+
"step": 17400
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.0488462691039857,
|
| 1243 |
+
"grad_norm": 0.04264242947101593,
|
| 1244 |
+
"learning_rate": 5.2841873938667464e-06,
|
| 1245 |
+
"loss": 0.0971,
|
| 1246 |
+
"step": 17500
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.0548396763560084,
|
| 1250 |
+
"grad_norm": 0.31686630845069885,
|
| 1251 |
+
"learning_rate": 5.2508906869110655e-06,
|
| 1252 |
+
"loss": 0.0998,
|
| 1253 |
+
"step": 17600
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 1.060833083608031,
|
| 1257 |
+
"grad_norm": 0.01217142678797245,
|
| 1258 |
+
"learning_rate": 5.217593979955383e-06,
|
| 1259 |
+
"loss": 0.099,
|
| 1260 |
+
"step": 17700
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 1.066826490860054,
|
| 1264 |
+
"grad_norm": 0.020899010822176933,
|
| 1265 |
+
"learning_rate": 5.184297272999701e-06,
|
| 1266 |
+
"loss": 0.0998,
|
| 1267 |
+
"step": 17800
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 1.0728198981120767,
|
| 1271 |
+
"grad_norm": 0.06026081368327141,
|
| 1272 |
+
"learning_rate": 5.151000566044019e-06,
|
| 1273 |
+
"loss": 0.0981,
|
| 1274 |
+
"step": 17900
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.0788133053640996,
|
| 1278 |
+
"grad_norm": 0.040572892874479294,
|
| 1279 |
+
"learning_rate": 5.117703859088337e-06,
|
| 1280 |
+
"loss": 0.0982,
|
| 1281 |
+
"step": 18000
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.0848067126161223,
|
| 1285 |
+
"grad_norm": 0.018748696893453598,
|
| 1286 |
+
"learning_rate": 5.084407152132654e-06,
|
| 1287 |
+
"loss": 0.0989,
|
| 1288 |
+
"step": 18100
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.090800119868145,
|
| 1292 |
+
"grad_norm": 0.16863317787647247,
|
| 1293 |
+
"learning_rate": 5.051110445176972e-06,
|
| 1294 |
+
"loss": 0.0984,
|
| 1295 |
+
"step": 18200
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.0967935271201679,
|
| 1299 |
+
"grad_norm": 0.238959401845932,
|
| 1300 |
+
"learning_rate": 5.01781373822129e-06,
|
| 1301 |
+
"loss": 0.0977,
|
| 1302 |
+
"step": 18300
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 1.1027869343721906,
|
| 1306 |
+
"grad_norm": 0.05871168524026871,
|
| 1307 |
+
"learning_rate": 4.984517031265608e-06,
|
| 1308 |
+
"loss": 0.0981,
|
| 1309 |
+
"step": 18400
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 1.1087803416242135,
|
| 1313 |
+
"grad_norm": 0.07862433046102524,
|
| 1314 |
+
"learning_rate": 4.951220324309926e-06,
|
| 1315 |
+
"loss": 0.0991,
|
| 1316 |
+
"step": 18500
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 1.1147737488762361,
|
| 1320 |
+
"grad_norm": 0.033283356577157974,
|
| 1321 |
+
"learning_rate": 4.917923617354244e-06,
|
| 1322 |
+
"loss": 0.0981,
|
| 1323 |
+
"step": 18600
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 1.1207671561282588,
|
| 1327 |
+
"grad_norm": 0.07558907568454742,
|
| 1328 |
+
"learning_rate": 4.884626910398562e-06,
|
| 1329 |
+
"loss": 0.0986,
|
| 1330 |
+
"step": 18700
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 1.1267605633802817,
|
| 1334 |
+
"grad_norm": 0.045116860419511795,
|
| 1335 |
+
"learning_rate": 4.85133020344288e-06,
|
| 1336 |
+
"loss": 0.0989,
|
| 1337 |
+
"step": 18800
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 1.1327539706323044,
|
| 1341 |
+
"grad_norm": 0.2039572298526764,
|
| 1342 |
+
"learning_rate": 4.818033496487198e-06,
|
| 1343 |
+
"loss": 0.099,
|
| 1344 |
+
"step": 18900
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 1.1387473778843273,
|
| 1348 |
+
"grad_norm": 0.026863813400268555,
|
| 1349 |
+
"learning_rate": 4.784736789531515e-06,
|
| 1350 |
+
"loss": 0.0995,
|
| 1351 |
+
"step": 19000
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 1.14474078513635,
|
| 1355 |
+
"grad_norm": 0.08295423537492752,
|
| 1356 |
+
"learning_rate": 4.751440082575834e-06,
|
| 1357 |
+
"loss": 0.0993,
|
| 1358 |
+
"step": 19100
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 1.1507341923883727,
|
| 1362 |
+
"grad_norm": 0.012610992416739464,
|
| 1363 |
+
"learning_rate": 4.718143375620152e-06,
|
| 1364 |
+
"loss": 0.1,
|
| 1365 |
+
"step": 19200
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 1.1567275996403956,
|
| 1369 |
+
"grad_norm": 0.15242983400821686,
|
| 1370 |
+
"learning_rate": 4.684846668664469e-06,
|
| 1371 |
+
"loss": 0.0993,
|
| 1372 |
+
"step": 19300
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 1.1627210068924183,
|
| 1376 |
+
"grad_norm": 0.03405880555510521,
|
| 1377 |
+
"learning_rate": 4.6515499617087876e-06,
|
| 1378 |
+
"loss": 0.098,
|
| 1379 |
+
"step": 19400
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.168714414144441,
|
| 1383 |
+
"grad_norm": 0.09746590256690979,
|
| 1384 |
+
"learning_rate": 4.618253254753105e-06,
|
| 1385 |
+
"loss": 0.0994,
|
| 1386 |
+
"step": 19500
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 1.174707821396464,
|
| 1390 |
+
"grad_norm": 0.01837330497801304,
|
| 1391 |
+
"learning_rate": 4.584956547797423e-06,
|
| 1392 |
+
"loss": 0.0993,
|
| 1393 |
+
"step": 19600
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.1807012286484866,
|
| 1397 |
+
"grad_norm": 0.06496419757604599,
|
| 1398 |
+
"learning_rate": 4.5516598408417415e-06,
|
| 1399 |
+
"loss": 0.0992,
|
| 1400 |
+
"step": 19700
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.1866946359005095,
|
| 1404 |
+
"grad_norm": 0.014315762557089329,
|
| 1405 |
+
"learning_rate": 4.518363133886059e-06,
|
| 1406 |
+
"loss": 0.0988,
|
| 1407 |
+
"step": 19800
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 1.1926880431525322,
|
| 1411 |
+
"grad_norm": 0.05952730402350426,
|
| 1412 |
+
"learning_rate": 4.485066426930376e-06,
|
| 1413 |
+
"loss": 0.0985,
|
| 1414 |
+
"step": 19900
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 1.198681450404555,
|
| 1418 |
+
"grad_norm": 0.18996870517730713,
|
| 1419 |
+
"learning_rate": 4.4517697199746955e-06,
|
| 1420 |
+
"loss": 0.0985,
|
| 1421 |
+
"step": 20000
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 1.2046748576565778,
|
| 1425 |
+
"grad_norm": 0.05301952362060547,
|
| 1426 |
+
"learning_rate": 4.418473013019013e-06,
|
| 1427 |
+
"loss": 0.1,
|
| 1428 |
+
"step": 20100
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 1.2106682649086005,
|
| 1432 |
+
"grad_norm": 0.15663863718509674,
|
| 1433 |
+
"learning_rate": 4.38517630606333e-06,
|
| 1434 |
+
"loss": 0.0992,
|
| 1435 |
+
"step": 20200
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 1.2166616721606234,
|
| 1439 |
+
"grad_norm": 0.2653251588344574,
|
| 1440 |
+
"learning_rate": 4.3518795991076486e-06,
|
| 1441 |
+
"loss": 0.0982,
|
| 1442 |
+
"step": 20300
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 1.222655079412646,
|
| 1446 |
+
"grad_norm": 0.11596585810184479,
|
| 1447 |
+
"learning_rate": 4.318582892151967e-06,
|
| 1448 |
+
"loss": 0.0984,
|
| 1449 |
+
"step": 20400
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 1.2286484866646687,
|
| 1453 |
+
"grad_norm": 0.12228166311979294,
|
| 1454 |
+
"learning_rate": 4.285286185196284e-06,
|
| 1455 |
+
"loss": 0.0987,
|
| 1456 |
+
"step": 20500
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.2346418939166917,
|
| 1460 |
+
"grad_norm": 0.19990171492099762,
|
| 1461 |
+
"learning_rate": 4.2519894782406025e-06,
|
| 1462 |
+
"loss": 0.0982,
|
| 1463 |
+
"step": 20600
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 1.2406353011687143,
|
| 1467 |
+
"grad_norm": 0.13501687347888947,
|
| 1468 |
+
"learning_rate": 4.21869277128492e-06,
|
| 1469 |
+
"loss": 0.099,
|
| 1470 |
+
"step": 20700
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 1.2466287084207373,
|
| 1474 |
+
"grad_norm": 0.14756691455841064,
|
| 1475 |
+
"learning_rate": 4.185396064329238e-06,
|
| 1476 |
+
"loss": 0.0989,
|
| 1477 |
+
"step": 20800
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 1.25262211567276,
|
| 1481 |
+
"grad_norm": 0.04468319192528725,
|
| 1482 |
+
"learning_rate": 4.1520993573735565e-06,
|
| 1483 |
+
"loss": 0.099,
|
| 1484 |
+
"step": 20900
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 1.2586155229247828,
|
| 1488 |
+
"grad_norm": 0.026138102635741234,
|
| 1489 |
+
"learning_rate": 4.118802650417874e-06,
|
| 1490 |
+
"loss": 0.0986,
|
| 1491 |
+
"step": 21000
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.2646089301768055,
|
| 1495 |
+
"grad_norm": 0.22098702192306519,
|
| 1496 |
+
"learning_rate": 4.085505943462192e-06,
|
| 1497 |
+
"loss": 0.0981,
|
| 1498 |
+
"step": 21100
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.2706023374288282,
|
| 1502 |
+
"grad_norm": 0.027339546009898186,
|
| 1503 |
+
"learning_rate": 4.05220923650651e-06,
|
| 1504 |
+
"loss": 0.0992,
|
| 1505 |
+
"step": 21200
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.2765957446808511,
|
| 1509 |
+
"grad_norm": 0.01603206805884838,
|
| 1510 |
+
"learning_rate": 4.018912529550828e-06,
|
| 1511 |
+
"loss": 0.0988,
|
| 1512 |
+
"step": 21300
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.2825891519328738,
|
| 1516 |
+
"grad_norm": 0.008753624744713306,
|
| 1517 |
+
"learning_rate": 3.985615822595146e-06,
|
| 1518 |
+
"loss": 0.0984,
|
| 1519 |
+
"step": 21400
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.2885825591848965,
|
| 1523 |
+
"grad_norm": 0.019739218056201935,
|
| 1524 |
+
"learning_rate": 3.9523191156394635e-06,
|
| 1525 |
+
"loss": 0.098,
|
| 1526 |
+
"step": 21500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 1.2945759664369194,
|
| 1530 |
+
"grad_norm": 0.8729631304740906,
|
| 1531 |
+
"learning_rate": 3.919022408683781e-06,
|
| 1532 |
+
"loss": 0.0983,
|
| 1533 |
+
"step": 21600
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 1.300569373688942,
|
| 1537 |
+
"grad_norm": 0.06987379491329193,
|
| 1538 |
+
"learning_rate": 3.885725701728099e-06,
|
| 1539 |
+
"loss": 0.0987,
|
| 1540 |
+
"step": 21700
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 1.306562780940965,
|
| 1544 |
+
"grad_norm": 0.11344248056411743,
|
| 1545 |
+
"learning_rate": 3.8524289947724175e-06,
|
| 1546 |
+
"loss": 0.0991,
|
| 1547 |
+
"step": 21800
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 1.3125561881929877,
|
| 1551 |
+
"grad_norm": 0.08928582072257996,
|
| 1552 |
+
"learning_rate": 3.819132287816735e-06,
|
| 1553 |
+
"loss": 0.0974,
|
| 1554 |
+
"step": 21900
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 1.3185495954450106,
|
| 1558 |
+
"grad_norm": 0.08214009553194046,
|
| 1559 |
+
"learning_rate": 3.785835580861053e-06,
|
| 1560 |
+
"loss": 0.0988,
|
| 1561 |
+
"step": 22000
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.3245430026970333,
|
| 1565 |
+
"grad_norm": 0.026567021384835243,
|
| 1566 |
+
"learning_rate": 3.752538873905371e-06,
|
| 1567 |
+
"loss": 0.0983,
|
| 1568 |
+
"step": 22100
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 1.330536409949056,
|
| 1572 |
+
"grad_norm": 0.1904483139514923,
|
| 1573 |
+
"learning_rate": 3.7192421669496893e-06,
|
| 1574 |
+
"loss": 0.0984,
|
| 1575 |
+
"step": 22200
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 1.3365298172010789,
|
| 1579 |
+
"grad_norm": 0.15923745930194855,
|
| 1580 |
+
"learning_rate": 3.685945459994007e-06,
|
| 1581 |
+
"loss": 0.0982,
|
| 1582 |
+
"step": 22300
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 1.3425232244531016,
|
| 1586 |
+
"grad_norm": 0.1682073026895523,
|
| 1587 |
+
"learning_rate": 3.652648753038325e-06,
|
| 1588 |
+
"loss": 0.0975,
|
| 1589 |
+
"step": 22400
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 1.3485166317051243,
|
| 1593 |
+
"grad_norm": 0.1236729547381401,
|
| 1594 |
+
"learning_rate": 3.6193520460826424e-06,
|
| 1595 |
+
"loss": 0.0986,
|
| 1596 |
+
"step": 22500
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 1.3545100389571472,
|
| 1600 |
+
"grad_norm": 0.2880057096481323,
|
| 1601 |
+
"learning_rate": 3.586055339126961e-06,
|
| 1602 |
+
"loss": 0.0991,
|
| 1603 |
+
"step": 22600
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.3605034462091699,
|
| 1607 |
+
"grad_norm": 0.07628654688596725,
|
| 1608 |
+
"learning_rate": 3.5527586321712785e-06,
|
| 1609 |
+
"loss": 0.0993,
|
| 1610 |
+
"step": 22700
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 1.3664968534611928,
|
| 1614 |
+
"grad_norm": 0.0549270398914814,
|
| 1615 |
+
"learning_rate": 3.5194619252155963e-06,
|
| 1616 |
+
"loss": 0.0983,
|
| 1617 |
+
"step": 22800
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.3724902607132154,
|
| 1621 |
+
"grad_norm": 0.010687282308936119,
|
| 1622 |
+
"learning_rate": 3.486165218259914e-06,
|
| 1623 |
+
"loss": 0.0992,
|
| 1624 |
+
"step": 22900
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.3784836679652384,
|
| 1628 |
+
"grad_norm": 0.32308948040008545,
|
| 1629 |
+
"learning_rate": 3.452868511304232e-06,
|
| 1630 |
+
"loss": 0.0985,
|
| 1631 |
+
"step": 23000
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 1.384477075217261,
|
| 1635 |
+
"grad_norm": 0.10801893472671509,
|
| 1636 |
+
"learning_rate": 3.4195718043485503e-06,
|
| 1637 |
+
"loss": 0.0981,
|
| 1638 |
+
"step": 23100
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 1.3904704824692837,
|
| 1642 |
+
"grad_norm": 0.035334184765815735,
|
| 1643 |
+
"learning_rate": 3.386275097392868e-06,
|
| 1644 |
+
"loss": 0.0981,
|
| 1645 |
+
"step": 23200
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 1.3964638897213066,
|
| 1649 |
+
"grad_norm": 0.3928769826889038,
|
| 1650 |
+
"learning_rate": 3.352978390437186e-06,
|
| 1651 |
+
"loss": 0.0987,
|
| 1652 |
+
"step": 23300
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 1.4024572969733293,
|
| 1656 |
+
"grad_norm": 0.03784007951617241,
|
| 1657 |
+
"learning_rate": 3.319681683481504e-06,
|
| 1658 |
+
"loss": 0.0983,
|
| 1659 |
+
"step": 23400
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 1.408450704225352,
|
| 1663 |
+
"grad_norm": 0.16313448548316956,
|
| 1664 |
+
"learning_rate": 3.286384976525822e-06,
|
| 1665 |
+
"loss": 0.0992,
|
| 1666 |
+
"step": 23500
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.414444111477375,
|
| 1670 |
+
"grad_norm": 0.7129063606262207,
|
| 1671 |
+
"learning_rate": 3.25308826957014e-06,
|
| 1672 |
+
"loss": 0.0985,
|
| 1673 |
+
"step": 23600
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 1.4204375187293976,
|
| 1677 |
+
"grad_norm": 0.08185921609401703,
|
| 1678 |
+
"learning_rate": 3.2197915626144578e-06,
|
| 1679 |
+
"loss": 0.0985,
|
| 1680 |
+
"step": 23700
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 1.4264309259814205,
|
| 1684 |
+
"grad_norm": 0.07759444415569305,
|
| 1685 |
+
"learning_rate": 3.1864948556587756e-06,
|
| 1686 |
+
"loss": 0.0987,
|
| 1687 |
+
"step": 23800
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 1.4324243332334432,
|
| 1691 |
+
"grad_norm": 0.15154729783535004,
|
| 1692 |
+
"learning_rate": 3.1531981487030935e-06,
|
| 1693 |
+
"loss": 0.0996,
|
| 1694 |
+
"step": 23900
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.4384177404854661,
|
| 1698 |
+
"grad_norm": 0.10120932757854462,
|
| 1699 |
+
"learning_rate": 3.1199014417474117e-06,
|
| 1700 |
+
"loss": 0.0979,
|
| 1701 |
+
"step": 24000
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 1.4444111477374888,
|
| 1705 |
+
"grad_norm": 0.042903125286102295,
|
| 1706 |
+
"learning_rate": 3.0866047347917296e-06,
|
| 1707 |
+
"loss": 0.0988,
|
| 1708 |
+
"step": 24100
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 1.4504045549895115,
|
| 1712 |
+
"grad_norm": 0.05083679407835007,
|
| 1713 |
+
"learning_rate": 3.053308027836047e-06,
|
| 1714 |
+
"loss": 0.0979,
|
| 1715 |
+
"step": 24200
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 1.4563979622415344,
|
| 1719 |
+
"grad_norm": 0.055504605174064636,
|
| 1720 |
+
"learning_rate": 3.020011320880365e-06,
|
| 1721 |
+
"loss": 0.0974,
|
| 1722 |
+
"step": 24300
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 1.462391369493557,
|
| 1726 |
+
"grad_norm": 0.10161745548248291,
|
| 1727 |
+
"learning_rate": 2.9867146139246835e-06,
|
| 1728 |
+
"loss": 0.0985,
|
| 1729 |
+
"step": 24400
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.4683847767455798,
|
| 1733 |
+
"grad_norm": 0.17584453523159027,
|
| 1734 |
+
"learning_rate": 2.953417906969001e-06,
|
| 1735 |
+
"loss": 0.0992,
|
| 1736 |
+
"step": 24500
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 1.4743781839976027,
|
| 1740 |
+
"grad_norm": 0.028774550184607506,
|
| 1741 |
+
"learning_rate": 2.9201212000133188e-06,
|
| 1742 |
+
"loss": 0.0995,
|
| 1743 |
+
"step": 24600
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 1.4803715912496254,
|
| 1747 |
+
"grad_norm": 0.15757901966571808,
|
| 1748 |
+
"learning_rate": 2.8868244930576366e-06,
|
| 1749 |
+
"loss": 0.0984,
|
| 1750 |
+
"step": 24700
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 1.486364998501648,
|
| 1754 |
+
"grad_norm": 0.08022894710302353,
|
| 1755 |
+
"learning_rate": 2.853527786101955e-06,
|
| 1756 |
+
"loss": 0.0988,
|
| 1757 |
+
"step": 24800
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 1.492358405753671,
|
| 1761 |
+
"grad_norm": 0.023040054365992546,
|
| 1762 |
+
"learning_rate": 2.8202310791462727e-06,
|
| 1763 |
+
"loss": 0.0982,
|
| 1764 |
+
"step": 24900
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 1.4983518130056939,
|
| 1768 |
+
"grad_norm": 0.032170914113521576,
|
| 1769 |
+
"learning_rate": 2.7869343721905906e-06,
|
| 1770 |
+
"loss": 0.0983,
|
| 1771 |
+
"step": 25000
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.5043452202577166,
|
| 1775 |
+
"grad_norm": 0.04139495640993118,
|
| 1776 |
+
"learning_rate": 2.7536376652349084e-06,
|
| 1777 |
+
"loss": 0.0986,
|
| 1778 |
+
"step": 25100
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 1.5103386275097392,
|
| 1782 |
+
"grad_norm": 0.11350516229867935,
|
| 1783 |
+
"learning_rate": 2.7203409582792263e-06,
|
| 1784 |
+
"loss": 0.0986,
|
| 1785 |
+
"step": 25200
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 1.5163320347617621,
|
| 1789 |
+
"grad_norm": 0.20275545120239258,
|
| 1790 |
+
"learning_rate": 2.6870442513235445e-06,
|
| 1791 |
+
"loss": 0.0994,
|
| 1792 |
+
"step": 25300
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 1.5223254420137848,
|
| 1796 |
+
"grad_norm": 0.013372101821005344,
|
| 1797 |
+
"learning_rate": 2.6537475443678624e-06,
|
| 1798 |
+
"loss": 0.0982,
|
| 1799 |
+
"step": 25400
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 1.5283188492658075,
|
| 1803 |
+
"grad_norm": 0.21790750324726105,
|
| 1804 |
+
"learning_rate": 2.6204508374121802e-06,
|
| 1805 |
+
"loss": 0.098,
|
| 1806 |
+
"step": 25500
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 1.5343122565178304,
|
| 1810 |
+
"grad_norm": 0.1723981350660324,
|
| 1811 |
+
"learning_rate": 2.587154130456498e-06,
|
| 1812 |
+
"loss": 0.0981,
|
| 1813 |
+
"step": 25600
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 1.5403056637698531,
|
| 1817 |
+
"grad_norm": 0.14081990718841553,
|
| 1818 |
+
"learning_rate": 2.5538574235008163e-06,
|
| 1819 |
+
"loss": 0.0986,
|
| 1820 |
+
"step": 25700
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 1.5462990710218758,
|
| 1824 |
+
"grad_norm": 0.1418023258447647,
|
| 1825 |
+
"learning_rate": 2.520560716545134e-06,
|
| 1826 |
+
"loss": 0.0977,
|
| 1827 |
+
"step": 25800
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 1.5522924782738987,
|
| 1831 |
+
"grad_norm": 0.010309278964996338,
|
| 1832 |
+
"learning_rate": 2.487264009589452e-06,
|
| 1833 |
+
"loss": 0.0982,
|
| 1834 |
+
"step": 25900
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 1.5582858855259216,
|
| 1838 |
+
"grad_norm": 0.06128934770822525,
|
| 1839 |
+
"learning_rate": 2.4539673026337694e-06,
|
| 1840 |
+
"loss": 0.0973,
|
| 1841 |
+
"step": 26000
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 1.5642792927779443,
|
| 1845 |
+
"grad_norm": 0.1110377237200737,
|
| 1846 |
+
"learning_rate": 2.4206705956780877e-06,
|
| 1847 |
+
"loss": 0.0992,
|
| 1848 |
+
"step": 26100
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 1.570272700029967,
|
| 1852 |
+
"grad_norm": 0.2141459882259369,
|
| 1853 |
+
"learning_rate": 2.3873738887224055e-06,
|
| 1854 |
+
"loss": 0.0989,
|
| 1855 |
+
"step": 26200
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 1.57626610728199,
|
| 1859 |
+
"grad_norm": 0.12720057368278503,
|
| 1860 |
+
"learning_rate": 2.3540771817667234e-06,
|
| 1861 |
+
"loss": 0.0986,
|
| 1862 |
+
"step": 26300
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 1.5822595145340126,
|
| 1866 |
+
"grad_norm": 0.0766364112496376,
|
| 1867 |
+
"learning_rate": 2.3207804748110412e-06,
|
| 1868 |
+
"loss": 0.0993,
|
| 1869 |
+
"step": 26400
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 1.5882529217860353,
|
| 1873 |
+
"grad_norm": 0.3568774163722992,
|
| 1874 |
+
"learning_rate": 2.2874837678553595e-06,
|
| 1875 |
+
"loss": 0.0981,
|
| 1876 |
+
"step": 26500
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 1.5942463290380582,
|
| 1880 |
+
"grad_norm": 0.2806101143360138,
|
| 1881 |
+
"learning_rate": 2.2541870608996773e-06,
|
| 1882 |
+
"loss": 0.0987,
|
| 1883 |
+
"step": 26600
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.6002397362900809,
|
| 1887 |
+
"grad_norm": 0.09340783953666687,
|
| 1888 |
+
"learning_rate": 2.220890353943995e-06,
|
| 1889 |
+
"loss": 0.0982,
|
| 1890 |
+
"step": 26700
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 1.6062331435421036,
|
| 1894 |
+
"grad_norm": 0.027837200090289116,
|
| 1895 |
+
"learning_rate": 2.187593646988313e-06,
|
| 1896 |
+
"loss": 0.0968,
|
| 1897 |
+
"step": 26800
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 1.6122265507941265,
|
| 1901 |
+
"grad_norm": 0.12748104333877563,
|
| 1902 |
+
"learning_rate": 2.154296940032631e-06,
|
| 1903 |
+
"loss": 0.0983,
|
| 1904 |
+
"step": 26900
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 1.6182199580461494,
|
| 1908 |
+
"grad_norm": 0.23494853079319,
|
| 1909 |
+
"learning_rate": 2.1210002330769487e-06,
|
| 1910 |
+
"loss": 0.0984,
|
| 1911 |
+
"step": 27000
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 1.624213365298172,
|
| 1915 |
+
"grad_norm": 0.18360479176044464,
|
| 1916 |
+
"learning_rate": 2.0877035261212665e-06,
|
| 1917 |
+
"loss": 0.0989,
|
| 1918 |
+
"step": 27100
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 1.6302067725501947,
|
| 1922 |
+
"grad_norm": 0.08933253586292267,
|
| 1923 |
+
"learning_rate": 2.054406819165585e-06,
|
| 1924 |
+
"loss": 0.0988,
|
| 1925 |
+
"step": 27200
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 1.6362001798022177,
|
| 1929 |
+
"grad_norm": 0.01740170828998089,
|
| 1930 |
+
"learning_rate": 2.0211101122099027e-06,
|
| 1931 |
+
"loss": 0.0997,
|
| 1932 |
+
"step": 27300
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"epoch": 1.6421935870542403,
|
| 1936 |
+
"grad_norm": 0.0467255637049675,
|
| 1937 |
+
"learning_rate": 1.9878134052542205e-06,
|
| 1938 |
+
"loss": 0.0975,
|
| 1939 |
+
"step": 27400
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 1.648186994306263,
|
| 1943 |
+
"grad_norm": 0.008605794049799442,
|
| 1944 |
+
"learning_rate": 1.9545166982985383e-06,
|
| 1945 |
+
"loss": 0.0989,
|
| 1946 |
+
"step": 27500
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 1.654180401558286,
|
| 1950 |
+
"grad_norm": 0.14202085137367249,
|
| 1951 |
+
"learning_rate": 1.9212199913428566e-06,
|
| 1952 |
+
"loss": 0.0992,
|
| 1953 |
+
"step": 27600
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 1.6601738088103086,
|
| 1957 |
+
"grad_norm": 0.09536932408809662,
|
| 1958 |
+
"learning_rate": 1.8879232843871742e-06,
|
| 1959 |
+
"loss": 0.0981,
|
| 1960 |
+
"step": 27700
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 1.6661672160623313,
|
| 1964 |
+
"grad_norm": 0.09608137607574463,
|
| 1965 |
+
"learning_rate": 1.8546265774314923e-06,
|
| 1966 |
+
"loss": 0.0986,
|
| 1967 |
+
"step": 27800
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 1.6721606233143542,
|
| 1971 |
+
"grad_norm": 0.022448230534791946,
|
| 1972 |
+
"learning_rate": 1.8213298704758101e-06,
|
| 1973 |
+
"loss": 0.0978,
|
| 1974 |
+
"step": 27900
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 1.6781540305663771,
|
| 1978 |
+
"grad_norm": 0.018090438097715378,
|
| 1979 |
+
"learning_rate": 1.7880331635201282e-06,
|
| 1980 |
+
"loss": 0.0986,
|
| 1981 |
+
"step": 28000
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 1.6841474378183996,
|
| 1985 |
+
"grad_norm": 0.016484877094626427,
|
| 1986 |
+
"learning_rate": 1.7547364565644458e-06,
|
| 1987 |
+
"loss": 0.098,
|
| 1988 |
+
"step": 28100
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 1.6901408450704225,
|
| 1992 |
+
"grad_norm": 0.024596238508820534,
|
| 1993 |
+
"learning_rate": 1.7214397496087637e-06,
|
| 1994 |
+
"loss": 0.0986,
|
| 1995 |
+
"step": 28200
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 1.6961342523224454,
|
| 1999 |
+
"grad_norm": 0.06954281032085419,
|
| 2000 |
+
"learning_rate": 1.6881430426530817e-06,
|
| 2001 |
+
"loss": 0.097,
|
| 2002 |
+
"step": 28300
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 1.702127659574468,
|
| 2006 |
+
"grad_norm": 0.31834906339645386,
|
| 2007 |
+
"learning_rate": 1.6548463356973996e-06,
|
| 2008 |
+
"loss": 0.0976,
|
| 2009 |
+
"step": 28400
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 1.7081210668264908,
|
| 2013 |
+
"grad_norm": 0.1658525913953781,
|
| 2014 |
+
"learning_rate": 1.6215496287417176e-06,
|
| 2015 |
+
"loss": 0.0974,
|
| 2016 |
+
"step": 28500
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 1.7141144740785137,
|
| 2020 |
+
"grad_norm": 0.030118102207779884,
|
| 2021 |
+
"learning_rate": 1.5882529217860355e-06,
|
| 2022 |
+
"loss": 0.0997,
|
| 2023 |
+
"step": 28600
|
| 2024 |
+
},
|
| 2025 |
+
{
|
| 2026 |
+
"epoch": 1.7201078813305364,
|
| 2027 |
+
"grad_norm": 0.16278527677059174,
|
| 2028 |
+
"learning_rate": 1.5549562148303535e-06,
|
| 2029 |
+
"loss": 0.098,
|
| 2030 |
+
"step": 28700
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 1.726101288582559,
|
| 2034 |
+
"grad_norm": 0.024548674002289772,
|
| 2035 |
+
"learning_rate": 1.5216595078746714e-06,
|
| 2036 |
+
"loss": 0.0986,
|
| 2037 |
+
"step": 28800
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 1.732094695834582,
|
| 2041 |
+
"grad_norm": 0.40551477670669556,
|
| 2042 |
+
"learning_rate": 1.4883628009189894e-06,
|
| 2043 |
+
"loss": 0.0972,
|
| 2044 |
+
"step": 28900
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 1.7380881030866049,
|
| 2048 |
+
"grad_norm": 0.05455062910914421,
|
| 2049 |
+
"learning_rate": 1.455066093963307e-06,
|
| 2050 |
+
"loss": 0.0976,
|
| 2051 |
+
"step": 29000
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 1.7440815103386273,
|
| 2055 |
+
"grad_norm": 0.455079585313797,
|
| 2056 |
+
"learning_rate": 1.421769387007625e-06,
|
| 2057 |
+
"loss": 0.0995,
|
| 2058 |
+
"step": 29100
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 1.7500749175906503,
|
| 2062 |
+
"grad_norm": 0.6706792116165161,
|
| 2063 |
+
"learning_rate": 1.388472680051943e-06,
|
| 2064 |
+
"loss": 0.0979,
|
| 2065 |
+
"step": 29200
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 1.7560683248426732,
|
| 2069 |
+
"grad_norm": 0.05078733339905739,
|
| 2070 |
+
"learning_rate": 1.3551759730962608e-06,
|
| 2071 |
+
"loss": 0.097,
|
| 2072 |
+
"step": 29300
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.7620617320946959,
|
| 2076 |
+
"grad_norm": 0.25584468245506287,
|
| 2077 |
+
"learning_rate": 1.3218792661405788e-06,
|
| 2078 |
+
"loss": 0.0981,
|
| 2079 |
+
"step": 29400
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 1.7680551393467185,
|
| 2083 |
+
"grad_norm": 0.12247732281684875,
|
| 2084 |
+
"learning_rate": 1.2885825591848967e-06,
|
| 2085 |
+
"loss": 0.0986,
|
| 2086 |
+
"step": 29500
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 1.7740485465987414,
|
| 2090 |
+
"grad_norm": 0.016069240868091583,
|
| 2091 |
+
"learning_rate": 1.2552858522292147e-06,
|
| 2092 |
+
"loss": 0.0987,
|
| 2093 |
+
"step": 29600
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 1.7800419538507641,
|
| 2097 |
+
"grad_norm": 0.1197570264339447,
|
| 2098 |
+
"learning_rate": 1.2219891452735326e-06,
|
| 2099 |
+
"loss": 0.0981,
|
| 2100 |
+
"step": 29700
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 1.7860353611027868,
|
| 2104 |
+
"grad_norm": 0.035120606422424316,
|
| 2105 |
+
"learning_rate": 1.1886924383178504e-06,
|
| 2106 |
+
"loss": 0.099,
|
| 2107 |
+
"step": 29800
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 1.7920287683548097,
|
| 2111 |
+
"grad_norm": 0.1425538957118988,
|
| 2112 |
+
"learning_rate": 1.1553957313621683e-06,
|
| 2113 |
+
"loss": 0.0989,
|
| 2114 |
+
"step": 29900
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 1.7980221756068326,
|
| 2118 |
+
"grad_norm": 0.027721386402845383,
|
| 2119 |
+
"learning_rate": 1.1220990244064863e-06,
|
| 2120 |
+
"loss": 0.0989,
|
| 2121 |
+
"step": 30000
|
| 2122 |
+
}
|
| 2123 |
+
],
|
| 2124 |
+
"logging_steps": 100,
|
| 2125 |
+
"max_steps": 33370,
|
| 2126 |
+
"num_input_tokens_seen": 0,
|
| 2127 |
+
"num_train_epochs": 2,
|
| 2128 |
+
"save_steps": 10000,
|
| 2129 |
+
"stateful_callbacks": {
|
| 2130 |
+
"TrainerControl": {
|
| 2131 |
+
"args": {
|
| 2132 |
+
"should_epoch_stop": false,
|
| 2133 |
+
"should_evaluate": false,
|
| 2134 |
+
"should_log": false,
|
| 2135 |
+
"should_save": true,
|
| 2136 |
+
"should_training_stop": false
|
| 2137 |
+
},
|
| 2138 |
+
"attributes": {}
|
| 2139 |
+
}
|
| 2140 |
+
},
|
| 2141 |
+
"total_flos": 3.663876369096977e+17,
|
| 2142 |
+
"train_batch_size": 8,
|
| 2143 |
+
"trial_name": null,
|
| 2144 |
+
"trial_params": null
|
| 2145 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffc14be46e769cef583b95cb894fb5d83917fbcc47aa64e192f6fa10c0d79b23
|
| 3 |
+
size 5240
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|