Upload 9 files
Browse files- config.json +226 -0
- generation_config.json +7 -0
- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- model.safetensors.index.json +0 -0
- trainer_state.json +355 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/raid/raushan/si-7b",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaInterleaveForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"btadapter": false,
|
| 7 |
+
"btadapter_depth": 4,
|
| 8 |
+
"clip_post_pretrain": null,
|
| 9 |
+
"clip_weight": "/code/ppllava_v0/siglip-so400m-patch14-384",
|
| 10 |
+
"frame_shape": [
|
| 11 |
+
27,
|
| 12 |
+
27
|
| 13 |
+
],
|
| 14 |
+
"hidden_size": 3584,
|
| 15 |
+
"ignore_index": -100,
|
| 16 |
+
"image_grid_pinpoints": [
|
| 17 |
+
[
|
| 18 |
+
384,
|
| 19 |
+
384
|
| 20 |
+
],
|
| 21 |
+
[
|
| 22 |
+
384,
|
| 23 |
+
768
|
| 24 |
+
],
|
| 25 |
+
[
|
| 26 |
+
384,
|
| 27 |
+
1152
|
| 28 |
+
],
|
| 29 |
+
[
|
| 30 |
+
384,
|
| 31 |
+
1536
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
384,
|
| 35 |
+
1920
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
384,
|
| 39 |
+
2304
|
| 40 |
+
],
|
| 41 |
+
[
|
| 42 |
+
768,
|
| 43 |
+
384
|
| 44 |
+
],
|
| 45 |
+
[
|
| 46 |
+
768,
|
| 47 |
+
768
|
| 48 |
+
],
|
| 49 |
+
[
|
| 50 |
+
768,
|
| 51 |
+
1152
|
| 52 |
+
],
|
| 53 |
+
[
|
| 54 |
+
768,
|
| 55 |
+
1536
|
| 56 |
+
],
|
| 57 |
+
[
|
| 58 |
+
768,
|
| 59 |
+
1920
|
| 60 |
+
],
|
| 61 |
+
[
|
| 62 |
+
768,
|
| 63 |
+
2304
|
| 64 |
+
],
|
| 65 |
+
[
|
| 66 |
+
1152,
|
| 67 |
+
384
|
| 68 |
+
],
|
| 69 |
+
[
|
| 70 |
+
1152,
|
| 71 |
+
768
|
| 72 |
+
],
|
| 73 |
+
[
|
| 74 |
+
1152,
|
| 75 |
+
1152
|
| 76 |
+
],
|
| 77 |
+
[
|
| 78 |
+
1152,
|
| 79 |
+
1536
|
| 80 |
+
],
|
| 81 |
+
[
|
| 82 |
+
1152,
|
| 83 |
+
1920
|
| 84 |
+
],
|
| 85 |
+
[
|
| 86 |
+
1152,
|
| 87 |
+
2304
|
| 88 |
+
],
|
| 89 |
+
[
|
| 90 |
+
1536,
|
| 91 |
+
384
|
| 92 |
+
],
|
| 93 |
+
[
|
| 94 |
+
1536,
|
| 95 |
+
768
|
| 96 |
+
],
|
| 97 |
+
[
|
| 98 |
+
1536,
|
| 99 |
+
1152
|
| 100 |
+
],
|
| 101 |
+
[
|
| 102 |
+
1536,
|
| 103 |
+
1536
|
| 104 |
+
],
|
| 105 |
+
[
|
| 106 |
+
1536,
|
| 107 |
+
1920
|
| 108 |
+
],
|
| 109 |
+
[
|
| 110 |
+
1536,
|
| 111 |
+
2304
|
| 112 |
+
],
|
| 113 |
+
[
|
| 114 |
+
1920,
|
| 115 |
+
384
|
| 116 |
+
],
|
| 117 |
+
[
|
| 118 |
+
1920,
|
| 119 |
+
768
|
| 120 |
+
],
|
| 121 |
+
[
|
| 122 |
+
1920,
|
| 123 |
+
1152
|
| 124 |
+
],
|
| 125 |
+
[
|
| 126 |
+
1920,
|
| 127 |
+
1536
|
| 128 |
+
],
|
| 129 |
+
[
|
| 130 |
+
1920,
|
| 131 |
+
1920
|
| 132 |
+
],
|
| 133 |
+
[
|
| 134 |
+
1920,
|
| 135 |
+
2304
|
| 136 |
+
],
|
| 137 |
+
[
|
| 138 |
+
2304,
|
| 139 |
+
384
|
| 140 |
+
],
|
| 141 |
+
[
|
| 142 |
+
2304,
|
| 143 |
+
768
|
| 144 |
+
],
|
| 145 |
+
[
|
| 146 |
+
2304,
|
| 147 |
+
1152
|
| 148 |
+
],
|
| 149 |
+
[
|
| 150 |
+
2304,
|
| 151 |
+
1536
|
| 152 |
+
],
|
| 153 |
+
[
|
| 154 |
+
2304,
|
| 155 |
+
1920
|
| 156 |
+
],
|
| 157 |
+
[
|
| 158 |
+
2304,
|
| 159 |
+
2304
|
| 160 |
+
]
|
| 161 |
+
],
|
| 162 |
+
"image_pooling_kernel": [
|
| 163 |
+
1,
|
| 164 |
+
3,
|
| 165 |
+
3
|
| 166 |
+
],
|
| 167 |
+
"image_pooling_stride": [
|
| 168 |
+
1,
|
| 169 |
+
3,
|
| 170 |
+
3
|
| 171 |
+
],
|
| 172 |
+
"image_token_index": 151646,
|
| 173 |
+
"long_clip": true,
|
| 174 |
+
"max_T": 64,
|
| 175 |
+
"model_type": "llava_onevision",
|
| 176 |
+
"pad_token_id": 151643,
|
| 177 |
+
"pooling": "clipST_3d",
|
| 178 |
+
"pooling_kernel": [
|
| 179 |
+
2,
|
| 180 |
+
3,
|
| 181 |
+
3
|
| 182 |
+
],
|
| 183 |
+
"pooling_stride": [
|
| 184 |
+
2,
|
| 185 |
+
3,
|
| 186 |
+
3
|
| 187 |
+
],
|
| 188 |
+
"pooling_temp": 0.01,
|
| 189 |
+
"projector_hidden_act": "gelu",
|
| 190 |
+
"qwen": true,
|
| 191 |
+
"text_config": {
|
| 192 |
+
"_name_or_path": "Qwen/Qwen2-7B-Instruct",
|
| 193 |
+
"architectures": [
|
| 194 |
+
"Qwen2ForCausalLM"
|
| 195 |
+
],
|
| 196 |
+
"bos_token_id": 151643,
|
| 197 |
+
"eos_token_id": 151645,
|
| 198 |
+
"hidden_size": 3584,
|
| 199 |
+
"intermediate_size": 18944,
|
| 200 |
+
"model_type": "qwen2",
|
| 201 |
+
"num_attention_heads": 28,
|
| 202 |
+
"num_hidden_layers": 28,
|
| 203 |
+
"num_key_value_heads": 4,
|
| 204 |
+
"rope_theta": 1000000.0,
|
| 205 |
+
"torch_dtype": "bfloat16",
|
| 206 |
+
"vocab_size": 152128
|
| 207 |
+
},
|
| 208 |
+
"tie_word_embeddings": false,
|
| 209 |
+
"torch_dtype": "bfloat16",
|
| 210 |
+
"transformers_version": "4.45.2",
|
| 211 |
+
"use_image_newline_parameter": true,
|
| 212 |
+
"video_token_index": 151647,
|
| 213 |
+
"vision_aspect_ratio": "anyres_max_9",
|
| 214 |
+
"vision_config": {
|
| 215 |
+
"hidden_size": 1152,
|
| 216 |
+
"image_size": 384,
|
| 217 |
+
"intermediate_size": 4304,
|
| 218 |
+
"model_type": "siglip_vision_model",
|
| 219 |
+
"num_attention_heads": 16,
|
| 220 |
+
"num_hidden_layers": 26,
|
| 221 |
+
"patch_size": 14,
|
| 222 |
+
"vision_use_head": false
|
| 223 |
+
},
|
| 224 |
+
"vision_feature_layer": -1,
|
| 225 |
+
"vision_feature_select_strategy": "full"
|
| 226 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"eos_token_id": 151645,
|
| 5 |
+
"pad_token_id": 151643,
|
| 6 |
+
"transformers_version": "4.45.2"
|
| 7 |
+
}
|
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:154acb85edfd1ebf6782693746740a0de21480b0686a232e651ec2aa5b482d2a
|
| 3 |
+
size 4909741188
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd0319fd43af1929432c3175d09e0c1e6d003311d7d1b2b0798ec9592e0cba64
|
| 3 |
+
size 4991497768
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3a94b21329d2841285464c5f0270e84f4bc5d818338800538dc89b5dc42b8b0
|
| 3 |
+
size 4932752872
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d700986509fa5f7f9d931b20a7226976bf9591fa53cc67166a2117ead617495a
|
| 3 |
+
size 2158500536
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
trainer_state.json
ADDED
|
@@ -0,0 +1,355 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.999787007454739,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2347,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.021299254526091587,
|
| 13 |
+
"grad_norm": 1.8116117629273891,
|
| 14 |
+
"learning_rate": 7.042253521126761e-06,
|
| 15 |
+
"loss": 1.1672,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.042598509052183174,
|
| 20 |
+
"grad_norm": 1.7801325074284498,
|
| 21 |
+
"learning_rate": 9.99599471666643e-06,
|
| 22 |
+
"loss": 1.0639,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06389776357827476,
|
| 27 |
+
"grad_norm": 1.7302714965761716,
|
| 28 |
+
"learning_rate": 9.970302558343623e-06,
|
| 29 |
+
"loss": 1.0591,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08519701810436635,
|
| 34 |
+
"grad_norm": 1.5998384203663196,
|
| 35 |
+
"learning_rate": 9.920945445527302e-06,
|
| 36 |
+
"loss": 1.055,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10649627263045794,
|
| 41 |
+
"grad_norm": 1.6500912559924017,
|
| 42 |
+
"learning_rate": 9.848158380778183e-06,
|
| 43 |
+
"loss": 1.0419,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12779552715654952,
|
| 48 |
+
"grad_norm": 1.596481575421808,
|
| 49 |
+
"learning_rate": 9.75228792299379e-06,
|
| 50 |
+
"loss": 1.0383,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14909478168264112,
|
| 55 |
+
"grad_norm": 1.6526899468527811,
|
| 56 |
+
"learning_rate": 9.633790537347847e-06,
|
| 57 |
+
"loss": 1.0334,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1703940362087327,
|
| 62 |
+
"grad_norm": 1.6150272900402385,
|
| 63 |
+
"learning_rate": 9.493230421936143e-06,
|
| 64 |
+
"loss": 1.0515,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.19169329073482427,
|
| 69 |
+
"grad_norm": 1.7820720198756743,
|
| 70 |
+
"learning_rate": 9.331276821476838e-06,
|
| 71 |
+
"loss": 1.0268,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.21299254526091588,
|
| 76 |
+
"grad_norm": 1.6035392653474185,
|
| 77 |
+
"learning_rate": 9.148700840855383e-06,
|
| 78 |
+
"loss": 1.0325,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.23429179978700745,
|
| 83 |
+
"grad_norm": 1.8039846181943315,
|
| 84 |
+
"learning_rate": 8.94637177368566e-06,
|
| 85 |
+
"loss": 1.0338,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.25559105431309903,
|
| 90 |
+
"grad_norm": 1.6713279817736957,
|
| 91 |
+
"learning_rate": 8.72525296336801e-06,
|
| 92 |
+
"loss": 1.0195,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.27689030883919064,
|
| 97 |
+
"grad_norm": 1.683168266494625,
|
| 98 |
+
"learning_rate": 8.486397216350815e-06,
|
| 99 |
+
"loss": 1.0222,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.29818956336528224,
|
| 104 |
+
"grad_norm": 1.6916656722965304,
|
| 105 |
+
"learning_rate": 8.230941789434305e-06,
|
| 106 |
+
"loss": 1.0155,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3194888178913738,
|
| 111 |
+
"grad_norm": 1.7135930235778254,
|
| 112 |
+
"learning_rate": 7.960102974983407e-06,
|
| 113 |
+
"loss": 1.0259,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3407880724174654,
|
| 118 |
+
"grad_norm": 1.6801315844607865,
|
| 119 |
+
"learning_rate": 7.675170309830907e-06,
|
| 120 |
+
"loss": 1.0244,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.362087326943557,
|
| 125 |
+
"grad_norm": 1.6004816476301602,
|
| 126 |
+
"learning_rate": 7.377500435443875e-06,
|
| 127 |
+
"loss": 1.0196,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.38338658146964855,
|
| 132 |
+
"grad_norm": 1.560394824607186,
|
| 133 |
+
"learning_rate": 7.068510638586805e-06,
|
| 134 |
+
"loss": 1.0213,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.40468583599574015,
|
| 139 |
+
"grad_norm": 1.7739863675472445,
|
| 140 |
+
"learning_rate": 6.7496721032360715e-06,
|
| 141 |
+
"loss": 1.0093,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.42598509052183176,
|
| 146 |
+
"grad_norm": 1.5424575836895802,
|
| 147 |
+
"learning_rate": 6.4225029058752076e-06,
|
| 148 |
+
"loss": 1.004,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4472843450479233,
|
| 153 |
+
"grad_norm": 1.7341809762292035,
|
| 154 |
+
"learning_rate": 6.088560787522257e-06,
|
| 155 |
+
"loss": 1.0103,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4685835995740149,
|
| 160 |
+
"grad_norm": 1.712705790413237,
|
| 161 |
+
"learning_rate": 5.749435736903566e-06,
|
| 162 |
+
"loss": 1.0029,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4898828541001065,
|
| 167 |
+
"grad_norm": 1.634153426002941,
|
| 168 |
+
"learning_rate": 5.406742420087504e-06,
|
| 169 |
+
"loss": 1.0043,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5111821086261981,
|
| 174 |
+
"grad_norm": 1.621004209807033,
|
| 175 |
+
"learning_rate": 5.062112492622702e-06,
|
| 176 |
+
"loss": 1.004,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5324813631522897,
|
| 181 |
+
"grad_norm": 1.6860145774112278,
|
| 182 |
+
"learning_rate": 4.717186830784763e-06,
|
| 183 |
+
"loss": 0.9951,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5537806176783813,
|
| 188 |
+
"grad_norm": 1.6211493662048333,
|
| 189 |
+
"learning_rate": 4.37360771892061e-06,
|
| 190 |
+
"loss": 1.0013,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5750798722044729,
|
| 195 |
+
"grad_norm": 1.8249160060138945,
|
| 196 |
+
"learning_rate": 4.0330110300886206e-06,
|
| 197 |
+
"loss": 0.9986,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5963791267305645,
|
| 202 |
+
"grad_norm": 1.6686984774536926,
|
| 203 |
+
"learning_rate": 3.697018437224645e-06,
|
| 204 |
+
"loss": 0.9967,
|
| 205 |
+
"step": 1400
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.617678381256656,
|
| 209 |
+
"grad_norm": 1.645263948081402,
|
| 210 |
+
"learning_rate": 3.3672296919186444e-06,
|
| 211 |
+
"loss": 1.0029,
|
| 212 |
+
"step": 1450
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6389776357827476,
|
| 216 |
+
"grad_norm": 1.7334881644315554,
|
| 217 |
+
"learning_rate": 3.0452150075647745e-06,
|
| 218 |
+
"loss": 0.9794,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6602768903088392,
|
| 223 |
+
"grad_norm": 1.6958071495211982,
|
| 224 |
+
"learning_rate": 2.732507583150834e-06,
|
| 225 |
+
"loss": 0.9866,
|
| 226 |
+
"step": 1550
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6815761448349308,
|
| 230 |
+
"grad_norm": 1.629069436410619,
|
| 231 |
+
"learning_rate": 2.4305963032832697e-06,
|
| 232 |
+
"loss": 0.9875,
|
| 233 |
+
"step": 1600
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7028753993610224,
|
| 237 |
+
"grad_norm": 1.7526521548397989,
|
| 238 |
+
"learning_rate": 2.140918649204985e-06,
|
| 239 |
+
"loss": 0.988,
|
| 240 |
+
"step": 1650
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.724174653887114,
|
| 244 |
+
"grad_norm": 1.717145624435277,
|
| 245 |
+
"learning_rate": 1.8648538545584133e-06,
|
| 246 |
+
"loss": 0.9803,
|
| 247 |
+
"step": 1700
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7454739084132055,
|
| 251 |
+
"grad_norm": 1.6652219634436365,
|
| 252 |
+
"learning_rate": 1.6037163384811998e-06,
|
| 253 |
+
"loss": 0.9931,
|
| 254 |
+
"step": 1750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7667731629392971,
|
| 258 |
+
"grad_norm": 1.4903724798212907,
|
| 259 |
+
"learning_rate": 1.3587494473012542e-06,
|
| 260 |
+
"loss": 0.9831,
|
| 261 |
+
"step": 1800
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7880724174653887,
|
| 265 |
+
"grad_norm": 1.6698441280484966,
|
| 266 |
+
"learning_rate": 1.1311195346286702e-06,
|
| 267 |
+
"loss": 0.9811,
|
| 268 |
+
"step": 1850
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8093716719914803,
|
| 272 |
+
"grad_norm": 1.7730255653260494,
|
| 273 |
+
"learning_rate": 9.219104080308027e-07,
|
| 274 |
+
"loss": 0.9726,
|
| 275 |
+
"step": 1900
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8306709265175719,
|
| 279 |
+
"grad_norm": 1.471953815903398,
|
| 280 |
+
"learning_rate": 7.321181687313578e-07,
|
| 281 |
+
"loss": 0.9786,
|
| 282 |
+
"step": 1950
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8519701810436635,
|
| 286 |
+
"grad_norm": 1.5610577043390987,
|
| 287 |
+
"learning_rate": 5.626464689031197e-07,
|
| 288 |
+
"loss": 0.9807,
|
| 289 |
+
"step": 2000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.873269435569755,
|
| 293 |
+
"grad_norm": 1.8173195064149255,
|
| 294 |
+
"learning_rate": 4.1430220913559914e-07,
|
| 295 |
+
"loss": 0.974,
|
| 296 |
+
"step": 2050
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8945686900958466,
|
| 300 |
+
"grad_norm": 1.7107057022608567,
|
| 301 |
+
"learning_rate": 2.877916965631344e-07,
|
| 302 |
+
"loss": 0.9874,
|
| 303 |
+
"step": 2100
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9158679446219382,
|
| 307 |
+
"grad_norm": 1.7134293009529347,
|
| 308 |
+
"learning_rate": 1.837172819456301e-07,
|
| 309 |
+
"loss": 0.9778,
|
| 310 |
+
"step": 2150
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9371671991480298,
|
| 314 |
+
"grad_norm": 1.7486730313830365,
|
| 315 |
+
"learning_rate": 1.025744917137117e-07,
|
| 316 |
+
"loss": 0.9691,
|
| 317 |
+
"step": 2200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9584664536741214,
|
| 321 |
+
"grad_norm": 1.6373608571865919,
|
| 322 |
+
"learning_rate": 4.474966863339203e-08,
|
| 323 |
+
"loss": 0.9826,
|
| 324 |
+
"step": 2250
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.979765708200213,
|
| 328 |
+
"grad_norm": 1.7726783615561228,
|
| 329 |
+
"learning_rate": 1.0518132323694696e-08,
|
| 330 |
+
"loss": 0.9805,
|
| 331 |
+
"step": 2300
|
| 332 |
+
}
|
| 333 |
+
],
|
| 334 |
+
"logging_steps": 50,
|
| 335 |
+
"max_steps": 2347,
|
| 336 |
+
"num_input_tokens_seen": 0,
|
| 337 |
+
"num_train_epochs": 1,
|
| 338 |
+
"save_steps": 500,
|
| 339 |
+
"stateful_callbacks": {
|
| 340 |
+
"TrainerControl": {
|
| 341 |
+
"args": {
|
| 342 |
+
"should_epoch_stop": false,
|
| 343 |
+
"should_evaluate": false,
|
| 344 |
+
"should_log": false,
|
| 345 |
+
"should_save": true,
|
| 346 |
+
"should_training_stop": true
|
| 347 |
+
},
|
| 348 |
+
"attributes": {}
|
| 349 |
+
}
|
| 350 |
+
},
|
| 351 |
+
"total_flos": 3.374008149160755e+16,
|
| 352 |
+
"train_batch_size": 4,
|
| 353 |
+
"trial_name": null,
|
| 354 |
+
"trial_params": null
|
| 355 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a7bd89f8a56ddaba93630ce32857ac868aed7d3b67faa286990f02fc6b8c259
|
| 3 |
+
size 7352
|