Upload folder using huggingface_hub
Browse files- README.md +9 -0
- adapter_config.json +83 -0
- adapter_model.bin +3 -0
- config.json +192 -0
- non_lora_trainables.bin +3 -0
- preprocessor_config.json +20 -0
- trainer_state.json +714 -0
README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "q-future/one-align",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 256,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 128,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"model.layers.7.self_attn.q_proj",
|
| 18 |
+
"model.layers.4.self_attn.v_proj.multiway.1",
|
| 19 |
+
"model.layers.23.self_attn.q_proj",
|
| 20 |
+
"model.layers.5.self_attn.q_proj",
|
| 21 |
+
"model.layers.11.self_attn.v_proj.multiway.1",
|
| 22 |
+
"model.layers.27.self_attn.v_proj.multiway.1",
|
| 23 |
+
"model.layers.22.self_attn.v_proj.multiway.1",
|
| 24 |
+
"model.layers.15.self_attn.v_proj.multiway.1",
|
| 25 |
+
"model.layers.19.self_attn.q_proj",
|
| 26 |
+
"model.layers.10.self_attn.v_proj.multiway.1",
|
| 27 |
+
"model.layers.30.self_attn.q_proj",
|
| 28 |
+
"model.layers.2.self_attn.q_proj",
|
| 29 |
+
"model.layers.10.self_attn.q_proj",
|
| 30 |
+
"model.layers.29.self_attn.q_proj",
|
| 31 |
+
"model.layers.9.self_attn.q_proj",
|
| 32 |
+
"model.layers.1.self_attn.v_proj.multiway.1",
|
| 33 |
+
"model.layers.24.self_attn.q_proj",
|
| 34 |
+
"model.layers.16.self_attn.q_proj",
|
| 35 |
+
"model.layers.30.self_attn.v_proj.multiway.1",
|
| 36 |
+
"model.layers.25.self_attn.q_proj",
|
| 37 |
+
"model.layers.4.self_attn.q_proj",
|
| 38 |
+
"model.layers.17.self_attn.q_proj",
|
| 39 |
+
"model.layers.8.self_attn.q_proj",
|
| 40 |
+
"model.layers.3.self_attn.v_proj.multiway.1",
|
| 41 |
+
"model.layers.20.self_attn.v_proj.multiway.1",
|
| 42 |
+
"model.layers.0.self_attn.q_proj",
|
| 43 |
+
"model.layers.14.self_attn.v_proj.multiway.1",
|
| 44 |
+
"model.layers.26.self_attn.q_proj",
|
| 45 |
+
"model.layers.6.self_attn.q_proj",
|
| 46 |
+
"model.layers.22.self_attn.q_proj",
|
| 47 |
+
"model.layers.23.self_attn.v_proj.multiway.1",
|
| 48 |
+
"model.layers.26.self_attn.v_proj.multiway.1",
|
| 49 |
+
"model.layers.7.self_attn.v_proj.multiway.1",
|
| 50 |
+
"model.layers.28.self_attn.v_proj.multiway.1",
|
| 51 |
+
"model.layers.19.self_attn.v_proj.multiway.1",
|
| 52 |
+
"model.layers.13.self_attn.v_proj.multiway.1",
|
| 53 |
+
"model.layers.5.self_attn.v_proj.multiway.1",
|
| 54 |
+
"model.layers.12.self_attn.v_proj.multiway.1",
|
| 55 |
+
"model.layers.9.self_attn.v_proj.multiway.1",
|
| 56 |
+
"model.layers.17.self_attn.v_proj.multiway.1",
|
| 57 |
+
"model.layers.18.self_attn.v_proj.multiway.1",
|
| 58 |
+
"model.layers.21.self_attn.v_proj.multiway.1",
|
| 59 |
+
"model.layers.27.self_attn.q_proj",
|
| 60 |
+
"model.layers.3.self_attn.q_proj",
|
| 61 |
+
"model.layers.12.self_attn.q_proj",
|
| 62 |
+
"model.layers.18.self_attn.q_proj",
|
| 63 |
+
"model.layers.8.self_attn.v_proj.multiway.1",
|
| 64 |
+
"model.layers.25.self_attn.v_proj.multiway.1",
|
| 65 |
+
"model.layers.2.self_attn.v_proj.multiway.1",
|
| 66 |
+
"model.layers.15.self_attn.q_proj",
|
| 67 |
+
"model.layers.31.self_attn.q_proj",
|
| 68 |
+
"model.layers.11.self_attn.q_proj",
|
| 69 |
+
"model.layers.6.self_attn.v_proj.multiway.1",
|
| 70 |
+
"model.layers.28.self_attn.q_proj",
|
| 71 |
+
"model.layers.14.self_attn.q_proj",
|
| 72 |
+
"model.layers.29.self_attn.v_proj.multiway.1",
|
| 73 |
+
"model.layers.21.self_attn.q_proj",
|
| 74 |
+
"model.layers.13.self_attn.q_proj",
|
| 75 |
+
"model.layers.24.self_attn.v_proj.multiway.1",
|
| 76 |
+
"model.layers.31.self_attn.v_proj.multiway.1",
|
| 77 |
+
"model.layers.0.self_attn.v_proj.multiway.1",
|
| 78 |
+
"model.layers.16.self_attn.v_proj.multiway.1",
|
| 79 |
+
"model.layers.20.self_attn.q_proj",
|
| 80 |
+
"model.layers.1.self_attn.q_proj"
|
| 81 |
+
],
|
| 82 |
+
"task_type": "CAUSAL_LM"
|
| 83 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df0fdb568f21f3d86f6ede5c29421ed10b8c7a323bd88a89a890345581851b1b
|
| 3 |
+
size 134262669
|
config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "q-future/one-align",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MPLUGOwl2LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "q-future/one-align--configuration_mplug_owl2.MPLUGOwl2Config",
|
| 10 |
+
"AutoModel": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM",
|
| 11 |
+
"AutoModelForCausalLM": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM"
|
| 12 |
+
},
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"eos_token_id": 2,
|
| 15 |
+
"freeze_vision_model": false,
|
| 16 |
+
"hidden_act": "silu",
|
| 17 |
+
"hidden_size": 4096,
|
| 18 |
+
"image_aspect_ratio": "pad",
|
| 19 |
+
"image_grid_pinpoints": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 11008,
|
| 22 |
+
"max_position_embeddings": 2048,
|
| 23 |
+
"model_type": "mplug_owl2",
|
| 24 |
+
"num_attention_heads": 32,
|
| 25 |
+
"num_hidden_layers": 32,
|
| 26 |
+
"num_key_value_heads": 32,
|
| 27 |
+
"pretraining_tp": 1,
|
| 28 |
+
"rms_norm_eps": 1e-06,
|
| 29 |
+
"rope_scaling": null,
|
| 30 |
+
"rope_theta": 10000.0,
|
| 31 |
+
"tie_word_embeddings": false,
|
| 32 |
+
"torch_dtype": "bfloat16",
|
| 33 |
+
"transformers_version": "4.36.1",
|
| 34 |
+
"tune_visual_abstractor": true,
|
| 35 |
+
"use_cache": true,
|
| 36 |
+
"visual_abstractor_lr": 2e-05,
|
| 37 |
+
"visual_config": {
|
| 38 |
+
"visual_abstractor": {
|
| 39 |
+
"_name_or_path": "",
|
| 40 |
+
"add_cross_attention": false,
|
| 41 |
+
"architectures": null,
|
| 42 |
+
"attention_probs_dropout_prob": 0.0,
|
| 43 |
+
"bad_words_ids": null,
|
| 44 |
+
"begin_suppress_tokens": null,
|
| 45 |
+
"bos_token_id": null,
|
| 46 |
+
"chunk_size_feed_forward": 0,
|
| 47 |
+
"cross_attention_hidden_size": null,
|
| 48 |
+
"decoder_start_token_id": null,
|
| 49 |
+
"diversity_penalty": 0.0,
|
| 50 |
+
"do_sample": false,
|
| 51 |
+
"early_stopping": false,
|
| 52 |
+
"encoder_hidden_size": 1024,
|
| 53 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 54 |
+
"eos_token_id": null,
|
| 55 |
+
"exponential_decay_length_penalty": null,
|
| 56 |
+
"finetuning_task": null,
|
| 57 |
+
"forced_bos_token_id": null,
|
| 58 |
+
"forced_eos_token_id": null,
|
| 59 |
+
"grid_size": 32,
|
| 60 |
+
"hidden_size": 1024,
|
| 61 |
+
"id2label": {
|
| 62 |
+
"0": "LABEL_0",
|
| 63 |
+
"1": "LABEL_1"
|
| 64 |
+
},
|
| 65 |
+
"initializer_range": 0.02,
|
| 66 |
+
"intermediate_size": 2816,
|
| 67 |
+
"is_decoder": false,
|
| 68 |
+
"is_encoder_decoder": false,
|
| 69 |
+
"label2id": {
|
| 70 |
+
"LABEL_0": 0,
|
| 71 |
+
"LABEL_1": 1
|
| 72 |
+
},
|
| 73 |
+
"layer_norm_eps": 1e-06,
|
| 74 |
+
"length_penalty": 1.0,
|
| 75 |
+
"max_length": 20,
|
| 76 |
+
"min_length": 0,
|
| 77 |
+
"model_type": "mplug_owl_visual_abstract",
|
| 78 |
+
"no_repeat_ngram_size": 0,
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_beam_groups": 1,
|
| 81 |
+
"num_beams": 1,
|
| 82 |
+
"num_hidden_layers": 6,
|
| 83 |
+
"num_learnable_queries": 64,
|
| 84 |
+
"num_return_sequences": 1,
|
| 85 |
+
"output_attentions": false,
|
| 86 |
+
"output_hidden_states": false,
|
| 87 |
+
"output_scores": false,
|
| 88 |
+
"pad_token_id": null,
|
| 89 |
+
"prefix": null,
|
| 90 |
+
"problem_type": null,
|
| 91 |
+
"pruned_heads": {},
|
| 92 |
+
"remove_invalid_values": false,
|
| 93 |
+
"repetition_penalty": 1.0,
|
| 94 |
+
"return_dict": true,
|
| 95 |
+
"return_dict_in_generate": false,
|
| 96 |
+
"sep_token_id": null,
|
| 97 |
+
"suppress_tokens": null,
|
| 98 |
+
"task_specific_params": null,
|
| 99 |
+
"temperature": 1.0,
|
| 100 |
+
"tf_legacy_loss": false,
|
| 101 |
+
"tie_encoder_decoder": false,
|
| 102 |
+
"tie_word_embeddings": true,
|
| 103 |
+
"tokenizer_class": null,
|
| 104 |
+
"top_k": 50,
|
| 105 |
+
"top_p": 1.0,
|
| 106 |
+
"torch_dtype": null,
|
| 107 |
+
"torchscript": false,
|
| 108 |
+
"transformers_version": "4.28.1",
|
| 109 |
+
"typical_p": 1.0,
|
| 110 |
+
"use_bfloat16": false
|
| 111 |
+
},
|
| 112 |
+
"visual_model": {
|
| 113 |
+
"_name_or_path": "",
|
| 114 |
+
"add_cross_attention": false,
|
| 115 |
+
"architectures": null,
|
| 116 |
+
"attention_dropout": 0.0,
|
| 117 |
+
"bad_words_ids": null,
|
| 118 |
+
"begin_suppress_tokens": null,
|
| 119 |
+
"bos_token_id": null,
|
| 120 |
+
"chunk_size_feed_forward": 0,
|
| 121 |
+
"cross_attention_hidden_size": null,
|
| 122 |
+
"decoder_start_token_id": null,
|
| 123 |
+
"diversity_penalty": 0.0,
|
| 124 |
+
"do_sample": false,
|
| 125 |
+
"early_stopping": false,
|
| 126 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 127 |
+
"eos_token_id": null,
|
| 128 |
+
"exponential_decay_length_penalty": null,
|
| 129 |
+
"finetuning_task": null,
|
| 130 |
+
"forced_bos_token_id": null,
|
| 131 |
+
"forced_eos_token_id": null,
|
| 132 |
+
"hidden_act": "quick_gelu",
|
| 133 |
+
"hidden_size": 1024,
|
| 134 |
+
"id2label": {
|
| 135 |
+
"0": "LABEL_0",
|
| 136 |
+
"1": "LABEL_1"
|
| 137 |
+
},
|
| 138 |
+
"image_size": 448,
|
| 139 |
+
"initializer_factor": 1.0,
|
| 140 |
+
"initializer_range": 0.02,
|
| 141 |
+
"intermediate_size": 4096,
|
| 142 |
+
"is_decoder": false,
|
| 143 |
+
"is_encoder_decoder": false,
|
| 144 |
+
"label2id": {
|
| 145 |
+
"LABEL_0": 0,
|
| 146 |
+
"LABEL_1": 1
|
| 147 |
+
},
|
| 148 |
+
"layer_norm_eps": 1e-06,
|
| 149 |
+
"length_penalty": 1.0,
|
| 150 |
+
"max_length": 20,
|
| 151 |
+
"min_length": 0,
|
| 152 |
+
"model_type": "mplug_owl_vision_model",
|
| 153 |
+
"no_repeat_ngram_size": 0,
|
| 154 |
+
"num_attention_heads": 16,
|
| 155 |
+
"num_beam_groups": 1,
|
| 156 |
+
"num_beams": 1,
|
| 157 |
+
"num_channels": 3,
|
| 158 |
+
"num_hidden_layers": 24,
|
| 159 |
+
"num_return_sequences": 1,
|
| 160 |
+
"output_attentions": false,
|
| 161 |
+
"output_hidden_states": false,
|
| 162 |
+
"output_scores": false,
|
| 163 |
+
"pad_token_id": null,
|
| 164 |
+
"patch_size": 14,
|
| 165 |
+
"prefix": null,
|
| 166 |
+
"problem_type": null,
|
| 167 |
+
"projection_dim": 768,
|
| 168 |
+
"pruned_heads": {},
|
| 169 |
+
"remove_invalid_values": false,
|
| 170 |
+
"repetition_penalty": 1.0,
|
| 171 |
+
"return_dict": true,
|
| 172 |
+
"return_dict_in_generate": false,
|
| 173 |
+
"sep_token_id": null,
|
| 174 |
+
"suppress_tokens": null,
|
| 175 |
+
"task_specific_params": null,
|
| 176 |
+
"temperature": 1.0,
|
| 177 |
+
"tf_legacy_loss": false,
|
| 178 |
+
"tie_encoder_decoder": false,
|
| 179 |
+
"tie_word_embeddings": true,
|
| 180 |
+
"tokenizer_class": null,
|
| 181 |
+
"top_k": 50,
|
| 182 |
+
"top_p": 1.0,
|
| 183 |
+
"torch_dtype": null,
|
| 184 |
+
"torchscript": false,
|
| 185 |
+
"transformers_version": "4.28.1",
|
| 186 |
+
"typical_p": 1.0,
|
| 187 |
+
"use_bfloat16": false,
|
| 188 |
+
"use_flash_attn": false
|
| 189 |
+
}
|
| 190 |
+
},
|
| 191 |
+
"vocab_size": 32000
|
| 192 |
+
}
|
non_lora_trainables.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:062a3c015cec832951d6130ab682f879441fb397eeecde135a836878087d6e83
|
| 3 |
+
size 162998185
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": 448,
|
| 3 |
+
"do_center_crop": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"feature_extractor_type": "CLIPFeatureExtractor",
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.48145466,
|
| 9 |
+
0.4578275,
|
| 10 |
+
0.40821073
|
| 11 |
+
],
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.26862954,
|
| 14 |
+
0.26130258,
|
| 15 |
+
0.27577711
|
| 16 |
+
],
|
| 17 |
+
"resample": 3,
|
| 18 |
+
"size": 448
|
| 19 |
+
}
|
| 20 |
+
|
trainer_state.json
ADDED
|
@@ -0,0 +1,714 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 114,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 5e-05,
|
| 14 |
+
"loss": 0.1289,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.05,
|
| 19 |
+
"learning_rate": 0.0001,
|
| 20 |
+
"loss": 0.1311,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.08,
|
| 25 |
+
"learning_rate": 0.00015000000000000001,
|
| 26 |
+
"loss": 0.1223,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.11,
|
| 31 |
+
"learning_rate": 0.0002,
|
| 32 |
+
"loss": 0.1101,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.13,
|
| 37 |
+
"learning_rate": 0.00019995921928281894,
|
| 38 |
+
"loss": 0.1267,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.16,
|
| 43 |
+
"learning_rate": 0.00019983691039261357,
|
| 44 |
+
"loss": 0.1282,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.18,
|
| 49 |
+
"learning_rate": 0.00019963317308626914,
|
| 50 |
+
"loss": 0.1038,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.21,
|
| 55 |
+
"learning_rate": 0.00019934817353485501,
|
| 56 |
+
"loss": 0.127,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.24,
|
| 61 |
+
"learning_rate": 0.0001989821441880933,
|
| 62 |
+
"loss": 0.094,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.26,
|
| 67 |
+
"learning_rate": 0.00019853538358476932,
|
| 68 |
+
"loss": 0.0962,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.29,
|
| 73 |
+
"learning_rate": 0.00019800825610923934,
|
| 74 |
+
"loss": 0.0842,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.32,
|
| 79 |
+
"learning_rate": 0.00019740119169423337,
|
| 80 |
+
"loss": 0.1035,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.34,
|
| 85 |
+
"learning_rate": 0.00019671468547019573,
|
| 86 |
+
"loss": 0.0779,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.37,
|
| 91 |
+
"learning_rate": 0.00019594929736144976,
|
| 92 |
+
"loss": 0.0685,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.39,
|
| 97 |
+
"learning_rate": 0.00019510565162951537,
|
| 98 |
+
"loss": 0.0872,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.42,
|
| 103 |
+
"learning_rate": 0.00019418443636395248,
|
| 104 |
+
"loss": 0.0964,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.45,
|
| 109 |
+
"learning_rate": 0.00019318640292114524,
|
| 110 |
+
"loss": 0.1025,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.47,
|
| 115 |
+
"learning_rate": 0.000192112365311485,
|
| 116 |
+
"loss": 0.1233,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.5,
|
| 121 |
+
"learning_rate": 0.00019096319953545185,
|
| 122 |
+
"loss": 0.0674,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.53,
|
| 127 |
+
"learning_rate": 0.00018973984286913584,
|
| 128 |
+
"loss": 0.1079,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.55,
|
| 133 |
+
"learning_rate": 0.00018844329309978145,
|
| 134 |
+
"loss": 0.0891,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.58,
|
| 139 |
+
"learning_rate": 0.00018707460771197774,
|
| 140 |
+
"loss": 0.1057,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.61,
|
| 145 |
+
"learning_rate": 0.0001856349030251589,
|
| 146 |
+
"loss": 0.1094,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.63,
|
| 151 |
+
"learning_rate": 0.00018412535328311814,
|
| 152 |
+
"loss": 0.0945,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.66,
|
| 157 |
+
"learning_rate": 0.0001825471896962774,
|
| 158 |
+
"loss": 0.093,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.68,
|
| 163 |
+
"learning_rate": 0.00018090169943749476,
|
| 164 |
+
"loss": 0.1067,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.71,
|
| 169 |
+
"learning_rate": 0.00017919022459222752,
|
| 170 |
+
"loss": 0.0718,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.74,
|
| 175 |
+
"learning_rate": 0.00017741416106390826,
|
| 176 |
+
"loss": 0.0786,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.76,
|
| 181 |
+
"learning_rate": 0.00017557495743542585,
|
| 182 |
+
"loss": 0.0852,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.79,
|
| 187 |
+
"learning_rate": 0.0001736741137876405,
|
| 188 |
+
"loss": 0.0796,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.82,
|
| 193 |
+
"learning_rate": 0.00017171318047589637,
|
| 194 |
+
"loss": 0.1008,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.84,
|
| 199 |
+
"learning_rate": 0.00016969375686552937,
|
| 200 |
+
"loss": 0.0952,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.87,
|
| 205 |
+
"learning_rate": 0.00016761749002740193,
|
| 206 |
+
"loss": 0.0881,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.89,
|
| 211 |
+
"learning_rate": 0.00016548607339452853,
|
| 212 |
+
"loss": 0.0955,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.92,
|
| 217 |
+
"learning_rate": 0.00016330124538088705,
|
| 218 |
+
"loss": 0.0823,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.95,
|
| 223 |
+
"learning_rate": 0.00016106478796354382,
|
| 224 |
+
"loss": 0.1021,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.97,
|
| 229 |
+
"learning_rate": 0.00015877852522924732,
|
| 230 |
+
"loss": 0.1055,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0,
|
| 235 |
+
"learning_rate": 0.00015644432188667695,
|
| 236 |
+
"loss": 0.0823,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.03,
|
| 241 |
+
"learning_rate": 0.00015406408174555976,
|
| 242 |
+
"loss": 0.082,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.05,
|
| 247 |
+
"learning_rate": 0.0001516397461638962,
|
| 248 |
+
"loss": 0.0913,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.08,
|
| 253 |
+
"learning_rate": 0.0001491732924645604,
|
| 254 |
+
"loss": 0.0723,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.11,
|
| 259 |
+
"learning_rate": 0.00014666673232256738,
|
| 260 |
+
"loss": 0.0957,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.13,
|
| 265 |
+
"learning_rate": 0.00014412211012432212,
|
| 266 |
+
"loss": 0.0708,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.16,
|
| 271 |
+
"learning_rate": 0.00014154150130018866,
|
| 272 |
+
"loss": 0.0619,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.18,
|
| 277 |
+
"learning_rate": 0.00013892701063173918,
|
| 278 |
+
"loss": 0.0955,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.21,
|
| 283 |
+
"learning_rate": 0.0001362807705350641,
|
| 284 |
+
"loss": 0.0798,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.24,
|
| 289 |
+
"learning_rate": 0.00013360493932154302,
|
| 290 |
+
"loss": 0.0767,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.26,
|
| 295 |
+
"learning_rate": 0.00013090169943749476,
|
| 296 |
+
"loss": 0.0754,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.29,
|
| 301 |
+
"learning_rate": 0.00012817325568414297,
|
| 302 |
+
"loss": 0.0798,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.32,
|
| 307 |
+
"learning_rate": 0.00012542183341934872,
|
| 308 |
+
"loss": 0.0859,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.34,
|
| 313 |
+
"learning_rate": 0.00012264967674257646,
|
| 314 |
+
"loss": 0.0859,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.37,
|
| 319 |
+
"learning_rate": 0.00011985904666457455,
|
| 320 |
+
"loss": 0.0795,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.39,
|
| 325 |
+
"learning_rate": 0.0001170522192632624,
|
| 326 |
+
"loss": 0.0671,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.42,
|
| 331 |
+
"learning_rate": 0.00011423148382732853,
|
| 332 |
+
"loss": 0.0791,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.45,
|
| 337 |
+
"learning_rate": 0.00011139914098905406,
|
| 338 |
+
"loss": 0.085,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.47,
|
| 343 |
+
"learning_rate": 0.00010855750084788398,
|
| 344 |
+
"loss": 0.0637,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.5,
|
| 349 |
+
"learning_rate": 0.00010570888108627681,
|
| 350 |
+
"loss": 0.0713,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.53,
|
| 355 |
+
"learning_rate": 0.00010285560507936961,
|
| 356 |
+
"loss": 0.0646,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.55,
|
| 361 |
+
"learning_rate": 0.0001,
|
| 362 |
+
"loss": 0.0585,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.58,
|
| 367 |
+
"learning_rate": 9.71443949206304e-05,
|
| 368 |
+
"loss": 0.0757,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.61,
|
| 373 |
+
"learning_rate": 9.42911189137232e-05,
|
| 374 |
+
"loss": 0.0551,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.63,
|
| 379 |
+
"learning_rate": 9.144249915211605e-05,
|
| 380 |
+
"loss": 0.0801,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.66,
|
| 385 |
+
"learning_rate": 8.860085901094595e-05,
|
| 386 |
+
"loss": 0.062,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.68,
|
| 391 |
+
"learning_rate": 8.57685161726715e-05,
|
| 392 |
+
"loss": 0.0614,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.71,
|
| 397 |
+
"learning_rate": 8.294778073673762e-05,
|
| 398 |
+
"loss": 0.0674,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.74,
|
| 403 |
+
"learning_rate": 8.014095333542548e-05,
|
| 404 |
+
"loss": 0.0803,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.76,
|
| 409 |
+
"learning_rate": 7.735032325742355e-05,
|
| 410 |
+
"loss": 0.0724,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.79,
|
| 415 |
+
"learning_rate": 7.457816658065134e-05,
|
| 416 |
+
"loss": 0.0701,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.82,
|
| 421 |
+
"learning_rate": 7.182674431585704e-05,
|
| 422 |
+
"loss": 0.082,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.84,
|
| 427 |
+
"learning_rate": 6.909830056250527e-05,
|
| 428 |
+
"loss": 0.0754,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.87,
|
| 433 |
+
"learning_rate": 6.639506067845697e-05,
|
| 434 |
+
"loss": 0.0789,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.89,
|
| 439 |
+
"learning_rate": 6.371922946493591e-05,
|
| 440 |
+
"loss": 0.0654,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.92,
|
| 445 |
+
"learning_rate": 6.107298936826086e-05,
|
| 446 |
+
"loss": 0.0616,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.95,
|
| 451 |
+
"learning_rate": 5.845849869981137e-05,
|
| 452 |
+
"loss": 0.0631,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.97,
|
| 457 |
+
"learning_rate": 5.5877889875677845e-05,
|
| 458 |
+
"loss": 0.0645,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.0,
|
| 463 |
+
"learning_rate": 5.333326767743263e-05,
|
| 464 |
+
"loss": 0.0596,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.03,
|
| 469 |
+
"learning_rate": 5.082670753543961e-05,
|
| 470 |
+
"loss": 0.0803,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.05,
|
| 475 |
+
"learning_rate": 4.836025383610382e-05,
|
| 476 |
+
"loss": 0.0559,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.08,
|
| 481 |
+
"learning_rate": 4.593591825444028e-05,
|
| 482 |
+
"loss": 0.0592,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.11,
|
| 487 |
+
"learning_rate": 4.355567811332311e-05,
|
| 488 |
+
"loss": 0.0509,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.13,
|
| 493 |
+
"learning_rate": 4.12214747707527e-05,
|
| 494 |
+
"loss": 0.0535,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.16,
|
| 499 |
+
"learning_rate": 3.893521203645618e-05,
|
| 500 |
+
"loss": 0.0558,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.18,
|
| 505 |
+
"learning_rate": 3.669875461911297e-05,
|
| 506 |
+
"loss": 0.0752,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 2.21,
|
| 511 |
+
"learning_rate": 3.45139266054715e-05,
|
| 512 |
+
"loss": 0.0551,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.24,
|
| 517 |
+
"learning_rate": 3.238250997259808e-05,
|
| 518 |
+
"loss": 0.0764,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.26,
|
| 523 |
+
"learning_rate": 3.030624313447067e-05,
|
| 524 |
+
"loss": 0.0531,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.29,
|
| 529 |
+
"learning_rate": 2.828681952410366e-05,
|
| 530 |
+
"loss": 0.0389,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 2.32,
|
| 535 |
+
"learning_rate": 2.6325886212359498e-05,
|
| 536 |
+
"loss": 0.0446,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 2.34,
|
| 541 |
+
"learning_rate": 2.4425042564574184e-05,
|
| 542 |
+
"loss": 0.0441,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.37,
|
| 547 |
+
"learning_rate": 2.2585838936091754e-05,
|
| 548 |
+
"loss": 0.0521,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.39,
|
| 553 |
+
"learning_rate": 2.0809775407772503e-05,
|
| 554 |
+
"loss": 0.064,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.42,
|
| 559 |
+
"learning_rate": 1.9098300562505266e-05,
|
| 560 |
+
"loss": 0.0491,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.45,
|
| 565 |
+
"learning_rate": 1.74528103037226e-05,
|
| 566 |
+
"loss": 0.0623,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.47,
|
| 571 |
+
"learning_rate": 1.587464671688187e-05,
|
| 572 |
+
"loss": 0.0574,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.5,
|
| 577 |
+
"learning_rate": 1.4365096974841108e-05,
|
| 578 |
+
"loss": 0.0432,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.53,
|
| 583 |
+
"learning_rate": 1.2925392288022298e-05,
|
| 584 |
+
"loss": 0.0562,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.55,
|
| 589 |
+
"learning_rate": 1.1556706900218572e-05,
|
| 590 |
+
"loss": 0.0527,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.58,
|
| 595 |
+
"learning_rate": 1.026015713086418e-05,
|
| 596 |
+
"loss": 0.0469,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.61,
|
| 601 |
+
"learning_rate": 9.036800464548157e-06,
|
| 602 |
+
"loss": 0.0497,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.63,
|
| 607 |
+
"learning_rate": 7.887634688515e-06,
|
| 608 |
+
"loss": 0.0592,
|
| 609 |
+
"step": 100
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.66,
|
| 613 |
+
"learning_rate": 6.813597078854772e-06,
|
| 614 |
+
"loss": 0.0491,
|
| 615 |
+
"step": 101
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.68,
|
| 619 |
+
"learning_rate": 5.8155636360475385e-06,
|
| 620 |
+
"loss": 0.0469,
|
| 621 |
+
"step": 102
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.71,
|
| 625 |
+
"learning_rate": 4.8943483704846475e-06,
|
| 626 |
+
"loss": 0.056,
|
| 627 |
+
"step": 103
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.74,
|
| 631 |
+
"learning_rate": 4.050702638550275e-06,
|
| 632 |
+
"loss": 0.0433,
|
| 633 |
+
"step": 104
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.76,
|
| 637 |
+
"learning_rate": 3.2853145298042953e-06,
|
| 638 |
+
"loss": 0.0514,
|
| 639 |
+
"step": 105
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.79,
|
| 643 |
+
"learning_rate": 2.5988083057666533e-06,
|
| 644 |
+
"loss": 0.0431,
|
| 645 |
+
"step": 106
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 2.82,
|
| 649 |
+
"learning_rate": 1.9917438907606556e-06,
|
| 650 |
+
"loss": 0.0609,
|
| 651 |
+
"step": 107
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 2.84,
|
| 655 |
+
"learning_rate": 1.4646164152307018e-06,
|
| 656 |
+
"loss": 0.0466,
|
| 657 |
+
"step": 108
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 2.87,
|
| 661 |
+
"learning_rate": 1.0178558119067315e-06,
|
| 662 |
+
"loss": 0.0697,
|
| 663 |
+
"step": 109
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.89,
|
| 667 |
+
"learning_rate": 6.518264651449779e-07,
|
| 668 |
+
"loss": 0.0427,
|
| 669 |
+
"step": 110
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.92,
|
| 673 |
+
"learning_rate": 3.6682691373086665e-07,
|
| 674 |
+
"loss": 0.0594,
|
| 675 |
+
"step": 111
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.95,
|
| 679 |
+
"learning_rate": 1.630896073864352e-07,
|
| 680 |
+
"loss": 0.0544,
|
| 681 |
+
"step": 112
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.97,
|
| 685 |
+
"learning_rate": 4.078071718107701e-08,
|
| 686 |
+
"loss": 0.0663,
|
| 687 |
+
"step": 113
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 3.0,
|
| 691 |
+
"learning_rate": 0.0,
|
| 692 |
+
"loss": 0.0521,
|
| 693 |
+
"step": 114
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 3.0,
|
| 697 |
+
"step": 114,
|
| 698 |
+
"total_flos": 2704210919424.0,
|
| 699 |
+
"train_loss": 0.07567342122395833,
|
| 700 |
+
"train_runtime": 1438.7994,
|
| 701 |
+
"train_samples_per_second": 4.967,
|
| 702 |
+
"train_steps_per_second": 0.079
|
| 703 |
+
}
|
| 704 |
+
],
|
| 705 |
+
"logging_steps": 1.0,
|
| 706 |
+
"max_steps": 114,
|
| 707 |
+
"num_input_tokens_seen": 0,
|
| 708 |
+
"num_train_epochs": 3,
|
| 709 |
+
"save_steps": 800,
|
| 710 |
+
"total_flos": 2704210919424.0,
|
| 711 |
+
"train_batch_size": 32,
|
| 712 |
+
"trial_name": null,
|
| 713 |
+
"trial_params": null
|
| 714 |
+
}
|