Upload folder using huggingface_hub
Browse files- README.md +9 -0
- adapter_config.json +26 -0
- adapter_model.bin +3 -0
- config.json +49 -0
- non_lora_trainables.bin +3 -0
- trainer_state.json +408 -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 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "liuhaotian/llava-v1.5-7b",
|
| 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 |
+
"k_proj",
|
| 18 |
+
"o_proj",
|
| 19 |
+
"down_proj",
|
| 20 |
+
"up_proj",
|
| 21 |
+
"q_proj",
|
| 22 |
+
"v_proj",
|
| 23 |
+
"gate_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa9859520d0dbc909f92889fd64a5b3bb8d2457c6c9a04512c1d7b7c505b0df3
|
| 3 |
+
size 639787082
|
config.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "liuhaotian/llava-v1.5-7b",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"freeze_mm_vision_resampler": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 4096,
|
| 14 |
+
"image_aspect_ratio": "pad",
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 11008,
|
| 17 |
+
"max_length": 4096,
|
| 18 |
+
"max_position_embeddings": 4096,
|
| 19 |
+
"mm_hidden_size": 1024,
|
| 20 |
+
"mm_patch_merge_type": "flat",
|
| 21 |
+
"mm_projector_lr": 2e-05,
|
| 22 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 23 |
+
"mm_resampler_type": null,
|
| 24 |
+
"mm_use_im_patch_token": false,
|
| 25 |
+
"mm_use_im_start_end": false,
|
| 26 |
+
"mm_vision_select_feature": "patch",
|
| 27 |
+
"mm_vision_select_layer": -2,
|
| 28 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 29 |
+
"model_type": "llava_llama",
|
| 30 |
+
"num_attention_heads": 32,
|
| 31 |
+
"num_hidden_layers": 32,
|
| 32 |
+
"num_key_value_heads": 32,
|
| 33 |
+
"pad_token_id": 0,
|
| 34 |
+
"pretraining_tp": 1,
|
| 35 |
+
"rms_norm_eps": 1e-05,
|
| 36 |
+
"rope_scaling": null,
|
| 37 |
+
"rope_theta": 10000.0,
|
| 38 |
+
"tie_word_embeddings": false,
|
| 39 |
+
"tokenizer_model_max_length": 2048,
|
| 40 |
+
"tokenizer_padding_side": "right",
|
| 41 |
+
"torch_dtype": "float16",
|
| 42 |
+
"transformers_version": "4.37.2",
|
| 43 |
+
"tune_mm_mlp_adapter": false,
|
| 44 |
+
"tune_mm_vision_resampler": false,
|
| 45 |
+
"unfreeze_mm_vision_tower": false,
|
| 46 |
+
"use_cache": true,
|
| 47 |
+
"use_mm_proj": true,
|
| 48 |
+
"vocab_size": 32000
|
| 49 |
+
}
|
non_lora_trainables.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff52ac35a68b2dbea4047f45ab8b48b8f6c98c08d6390e32d80d9142a9e8f3b2
|
| 3 |
+
size 41961648
|
trainer_state.json
ADDED
|
@@ -0,0 +1,408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 63,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"learning_rate": 0.0001,
|
| 14 |
+
"loss": 2.5627,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.1,
|
| 19 |
+
"learning_rate": 0.0002,
|
| 20 |
+
"loss": 2.5872,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.14,
|
| 25 |
+
"learning_rate": 0.00019986740898848306,
|
| 26 |
+
"loss": 0.2394,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.19,
|
| 31 |
+
"learning_rate": 0.0001994699875614589,
|
| 32 |
+
"loss": 0.3004,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.24,
|
| 37 |
+
"learning_rate": 0.00019880878960910772,
|
| 38 |
+
"loss": 0.2071,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.29,
|
| 43 |
+
"learning_rate": 0.0001978855685095358,
|
| 44 |
+
"loss": 0.2135,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.33,
|
| 49 |
+
"learning_rate": 0.00019670277247913205,
|
| 50 |
+
"loss": 0.1922,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.38,
|
| 55 |
+
"learning_rate": 0.00019526353808033825,
|
| 56 |
+
"loss": 0.211,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.43,
|
| 61 |
+
"learning_rate": 0.00019357168190404936,
|
| 62 |
+
"loss": 0.1895,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.48,
|
| 67 |
+
"learning_rate": 0.0001916316904487005,
|
| 68 |
+
"loss": 0.1772,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.52,
|
| 73 |
+
"learning_rate": 0.00018944870822287956,
|
| 74 |
+
"loss": 0.1635,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.57,
|
| 79 |
+
"learning_rate": 0.00018702852410301554,
|
| 80 |
+
"loss": 0.1521,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.62,
|
| 85 |
+
"learning_rate": 0.00018437755598231856,
|
| 86 |
+
"loss": 0.2165,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.67,
|
| 91 |
+
"learning_rate": 0.00018150283375168114,
|
| 92 |
+
"loss": 0.1462,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.71,
|
| 97 |
+
"learning_rate": 0.00017841198065767107,
|
| 98 |
+
"loss": 0.1535,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.76,
|
| 103 |
+
"learning_rate": 0.00017511319308705198,
|
| 104 |
+
"loss": 0.1654,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.81,
|
| 109 |
+
"learning_rate": 0.00017161521883143934,
|
| 110 |
+
"loss": 0.1542,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.86,
|
| 115 |
+
"learning_rate": 0.00016792733388972932,
|
| 116 |
+
"loss": 0.1482,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.9,
|
| 121 |
+
"learning_rate": 0.00016405931786981755,
|
| 122 |
+
"loss": 0.1389,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.95,
|
| 127 |
+
"learning_rate": 0.00016002142805483685,
|
| 128 |
+
"loss": 0.1234,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0,
|
| 133 |
+
"learning_rate": 0.00015582437220268647,
|
| 134 |
+
"loss": 0.1184,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.05,
|
| 139 |
+
"learning_rate": 0.0001514792801509831,
|
| 140 |
+
"loss": 0.11,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.1,
|
| 145 |
+
"learning_rate": 0.000146997674302732,
|
| 146 |
+
"loss": 0.1085,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.14,
|
| 151 |
+
"learning_rate": 0.0001423914390709861,
|
| 152 |
+
"loss": 0.108,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.19,
|
| 157 |
+
"learning_rate": 0.00013767278936351854,
|
| 158 |
+
"loss": 0.1044,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.24,
|
| 163 |
+
"learning_rate": 0.0001328542381910835,
|
| 164 |
+
"loss": 0.0888,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.29,
|
| 169 |
+
"learning_rate": 0.00012794856348516095,
|
| 170 |
+
"loss": 0.1162,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.33,
|
| 175 |
+
"learning_rate": 0.0001229687742131796,
|
| 176 |
+
"loss": 0.0989,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.38,
|
| 181 |
+
"learning_rate": 0.00011792807588107357,
|
| 182 |
+
"loss": 0.0849,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.43,
|
| 187 |
+
"learning_rate": 0.00011283983551465511,
|
| 188 |
+
"loss": 0.1005,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.48,
|
| 193 |
+
"learning_rate": 0.00010771754621266466,
|
| 194 |
+
"loss": 0.0979,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.52,
|
| 199 |
+
"learning_rate": 0.00010257479136549889,
|
| 200 |
+
"loss": 0.0917,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.57,
|
| 205 |
+
"learning_rate": 9.742520863450115e-05,
|
| 206 |
+
"loss": 0.0946,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.62,
|
| 211 |
+
"learning_rate": 9.228245378733537e-05,
|
| 212 |
+
"loss": 0.0871,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.67,
|
| 217 |
+
"learning_rate": 8.71601644853449e-05,
|
| 218 |
+
"loss": 0.0733,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.71,
|
| 223 |
+
"learning_rate": 8.207192411892646e-05,
|
| 224 |
+
"loss": 0.0797,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.76,
|
| 229 |
+
"learning_rate": 7.703122578682046e-05,
|
| 230 |
+
"loss": 0.0914,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.81,
|
| 235 |
+
"learning_rate": 7.205143651483906e-05,
|
| 236 |
+
"loss": 0.0562,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.86,
|
| 241 |
+
"learning_rate": 6.714576180891654e-05,
|
| 242 |
+
"loss": 0.0735,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.9,
|
| 247 |
+
"learning_rate": 6.232721063648148e-05,
|
| 248 |
+
"loss": 0.048,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.95,
|
| 253 |
+
"learning_rate": 5.7608560929013946e-05,
|
| 254 |
+
"loss": 0.0603,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.0,
|
| 259 |
+
"learning_rate": 5.300232569726804e-05,
|
| 260 |
+
"loss": 0.0856,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.05,
|
| 265 |
+
"learning_rate": 4.852071984901696e-05,
|
| 266 |
+
"loss": 0.0655,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.1,
|
| 271 |
+
"learning_rate": 4.417562779731355e-05,
|
| 272 |
+
"loss": 0.0494,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.14,
|
| 277 |
+
"learning_rate": 3.997857194516319e-05,
|
| 278 |
+
"loss": 0.0334,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.19,
|
| 283 |
+
"learning_rate": 3.594068213018249e-05,
|
| 284 |
+
"loss": 0.0378,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.24,
|
| 289 |
+
"learning_rate": 3.207266611027069e-05,
|
| 290 |
+
"loss": 0.0425,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.29,
|
| 295 |
+
"learning_rate": 2.8384781168560693e-05,
|
| 296 |
+
"loss": 0.0272,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.33,
|
| 301 |
+
"learning_rate": 2.4886806912948035e-05,
|
| 302 |
+
"loss": 0.0511,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.38,
|
| 307 |
+
"learning_rate": 2.1588019342328968e-05,
|
| 308 |
+
"loss": 0.0393,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.43,
|
| 313 |
+
"learning_rate": 1.8497166248318876e-05,
|
| 314 |
+
"loss": 0.0265,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.48,
|
| 319 |
+
"learning_rate": 1.562244401768144e-05,
|
| 320 |
+
"loss": 0.0442,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.52,
|
| 325 |
+
"learning_rate": 1.2971475896984475e-05,
|
| 326 |
+
"loss": 0.0196,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.57,
|
| 331 |
+
"learning_rate": 1.0551291777120464e-05,
|
| 332 |
+
"loss": 0.0296,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.62,
|
| 337 |
+
"learning_rate": 8.368309551299536e-06,
|
| 338 |
+
"loss": 0.0482,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.67,
|
| 343 |
+
"learning_rate": 6.428318095950647e-06,
|
| 344 |
+
"loss": 0.0442,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.71,
|
| 349 |
+
"learning_rate": 4.7364619196617495e-06,
|
| 350 |
+
"loss": 0.0442,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.76,
|
| 355 |
+
"learning_rate": 3.2972275208679625e-06,
|
| 356 |
+
"loss": 0.027,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.81,
|
| 361 |
+
"learning_rate": 2.1144314904642195e-06,
|
| 362 |
+
"loss": 0.053,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.86,
|
| 367 |
+
"learning_rate": 1.1912103908922945e-06,
|
| 368 |
+
"loss": 0.037,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.9,
|
| 373 |
+
"learning_rate": 5.300124385410943e-07,
|
| 374 |
+
"loss": 0.0439,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.95,
|
| 379 |
+
"learning_rate": 1.3259101151694708e-07,
|
| 380 |
+
"loss": 0.0234,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.0,
|
| 385 |
+
"learning_rate": 0.0,
|
| 386 |
+
"loss": 0.0241,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.0,
|
| 391 |
+
"step": 63,
|
| 392 |
+
"total_flos": 3251608289280.0,
|
| 393 |
+
"train_loss": 0.1782660081923481,
|
| 394 |
+
"train_runtime": 283.1757,
|
| 395 |
+
"train_samples_per_second": 27.947,
|
| 396 |
+
"train_steps_per_second": 0.222
|
| 397 |
+
}
|
| 398 |
+
],
|
| 399 |
+
"logging_steps": 1.0,
|
| 400 |
+
"max_steps": 63,
|
| 401 |
+
"num_input_tokens_seen": 0,
|
| 402 |
+
"num_train_epochs": 3,
|
| 403 |
+
"save_steps": 50000,
|
| 404 |
+
"total_flos": 3251608289280.0,
|
| 405 |
+
"train_batch_size": 16,
|
| 406 |
+
"trial_name": null,
|
| 407 |
+
"trial_params": null
|
| 408 |
+
}
|