End of training
Browse files- README.md +2 -2
- all_results.json +10 -10
- eval_layer_index_28_mat_mlp.c_proj_results.json +7 -7
- model-00001-of-00002.safetensors +2 -2
- model.safetensors.index.json +10 -4
- runs/Mar13_16-43-27_demo2/events.out.tfevents.1741939466.demo2.205134.4 +3 -0
- runs/Mar13_16-43-27_demo2/events.out.tfevents.1741953105.demo2.205134.5 +3 -0
- train_layer_index_28_mat_mlp.c_proj_results.json +6 -6
- trainer_state.json +592 -592
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
base_model: LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct
|
| 3 |
-
datasets:
|
| 4 |
library_name: transformers
|
| 5 |
model_name: output_exaone_tucker_rank16_newdata_b1_g2_gradient2_e1_b512
|
| 6 |
tags:
|
|
@@ -12,7 +12,7 @@ licence: license
|
|
| 12 |
|
| 13 |
# Model Card for output_exaone_tucker_rank16_newdata_b1_g2_gradient2_e1_b512
|
| 14 |
|
| 15 |
-
This model is a fine-tuned version of [LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct](https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct) on the [
|
| 16 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 17 |
|
| 18 |
## Quick start
|
|
|
|
| 1 |
---
|
| 2 |
base_model: LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct
|
| 3 |
+
datasets: testing_layer_28_Matrix_mlp.c_proj
|
| 4 |
library_name: transformers
|
| 5 |
model_name: output_exaone_tucker_rank16_newdata_b1_g2_gradient2_e1_b512
|
| 6 |
tags:
|
|
|
|
| 12 |
|
| 13 |
# Model Card for output_exaone_tucker_rank16_newdata_b1_g2_gradient2_e1_b512
|
| 14 |
|
| 15 |
+
This model is a fine-tuned version of [LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct](https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-2.4B-Instruct) on the [testing_layer_28_Matrix_mlp.c_proj](https://huggingface.co/datasets/testing_layer_28_Matrix_mlp.c_proj) dataset.
|
| 16 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 17 |
|
| 18 |
## Quick start
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"eval_accuracy": 0.
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 25971,
|
| 6 |
-
"eval_samples_per_second": 27.
|
| 7 |
-
"eval_steps_per_second": 13.
|
| 8 |
-
"perplexity": 5.
|
| 9 |
-
"total_flos": 9.
|
| 10 |
-
"train_loss": 1.
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 294393,
|
| 13 |
-
"train_samples_per_second": 23.
|
| 14 |
"train_steps_per_second": 5.806
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"eval_accuracy": 0.8091990942751283,
|
| 3 |
+
"eval_loss": 1.6708984375,
|
| 4 |
+
"eval_runtime": 934.1658,
|
| 5 |
"eval_samples": 25971,
|
| 6 |
+
"eval_samples_per_second": 27.801,
|
| 7 |
+
"eval_steps_per_second": 13.901,
|
| 8 |
+
"perplexity": 5.31694259254287,
|
| 9 |
+
"total_flos": 9.078073041438638e+17,
|
| 10 |
+
"train_loss": 1.6789688296530434,
|
| 11 |
+
"train_runtime": 12675.9905,
|
| 12 |
"train_samples": 294393,
|
| 13 |
+
"train_samples_per_second": 23.224,
|
| 14 |
"train_steps_per_second": 5.806
|
| 15 |
}
|
eval_layer_index_28_mat_mlp.c_proj_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"eval_accuracy": 0.
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 27.
|
| 7 |
-
"eval_steps_per_second": 13.
|
| 8 |
-
"perplexity":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"eval_accuracy": 0.8091990942751283,
|
| 3 |
+
"eval_loss": 1.6708984375,
|
| 4 |
+
"eval_runtime": 934.1658,
|
| 5 |
+
"eval_samples": 25971,
|
| 6 |
+
"eval_samples_per_second": 27.801,
|
| 7 |
+
"eval_steps_per_second": 13.901,
|
| 8 |
+
"perplexity": 5.31694259254287
|
| 9 |
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4a4e85ebd672ede1b25882491b132e29791e040f5587628d0261e4debb03cf3
|
| 3 |
+
size 4560124544
|
model.safetensors.index.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
-
"total_size":
|
| 4 |
},
|
| 5 |
"weight_map": {
|
| 6 |
"lm_head.weight": "model-00002-of-00002.safetensors",
|
|
@@ -199,9 +199,15 @@
|
|
| 199 |
"transformer.h.28.attn.attention.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 200 |
"transformer.h.28.ln_1.weight": "model-00001-of-00002.safetensors",
|
| 201 |
"transformer.h.28.ln_2.weight": "model-00001-of-00002.safetensors",
|
| 202 |
-
"transformer.h.28.mlp.c_fc_0.
|
| 203 |
-
"transformer.h.28.mlp.
|
| 204 |
-
"transformer.h.28.mlp.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 205 |
"transformer.h.29.attn.attention.k_proj.tucker_core": "model-00001-of-00002.safetensors",
|
| 206 |
"transformer.h.29.attn.attention.k_proj.tucker_factors.0": "model-00001-of-00002.safetensors",
|
| 207 |
"transformer.h.29.attn.attention.k_proj.tucker_factors.1": "model-00001-of-00002.safetensors",
|
|
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
+
"total_size": 5084379136
|
| 4 |
},
|
| 5 |
"weight_map": {
|
| 6 |
"lm_head.weight": "model-00002-of-00002.safetensors",
|
|
|
|
| 199 |
"transformer.h.28.attn.attention.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 200 |
"transformer.h.28.ln_1.weight": "model-00001-of-00002.safetensors",
|
| 201 |
"transformer.h.28.ln_2.weight": "model-00001-of-00002.safetensors",
|
| 202 |
+
"transformer.h.28.mlp.c_fc_0.tucker_core": "model-00001-of-00002.safetensors",
|
| 203 |
+
"transformer.h.28.mlp.c_fc_0.tucker_factors.0": "model-00001-of-00002.safetensors",
|
| 204 |
+
"transformer.h.28.mlp.c_fc_0.tucker_factors.1": "model-00001-of-00002.safetensors",
|
| 205 |
+
"transformer.h.28.mlp.c_fc_1.tucker_core": "model-00001-of-00002.safetensors",
|
| 206 |
+
"transformer.h.28.mlp.c_fc_1.tucker_factors.0": "model-00001-of-00002.safetensors",
|
| 207 |
+
"transformer.h.28.mlp.c_fc_1.tucker_factors.1": "model-00001-of-00002.safetensors",
|
| 208 |
+
"transformer.h.28.mlp.c_proj.tucker_core": "model-00001-of-00002.safetensors",
|
| 209 |
+
"transformer.h.28.mlp.c_proj.tucker_factors.0": "model-00001-of-00002.safetensors",
|
| 210 |
+
"transformer.h.28.mlp.c_proj.tucker_factors.1": "model-00001-of-00002.safetensors",
|
| 211 |
"transformer.h.29.attn.attention.k_proj.tucker_core": "model-00001-of-00002.safetensors",
|
| 212 |
"transformer.h.29.attn.attention.k_proj.tucker_factors.0": "model-00001-of-00002.safetensors",
|
| 213 |
"transformer.h.29.attn.attention.k_proj.tucker_factors.1": "model-00001-of-00002.safetensors",
|
runs/Mar13_16-43-27_demo2/events.out.tfevents.1741939466.demo2.205134.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a31e6a5ac85d31876f79f8c799c349c3350b3d66c113c5662158f670f5eec26
|
| 3 |
+
size 47428
|
runs/Mar13_16-43-27_demo2/events.out.tfevents.1741953105.demo2.205134.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57a790b4d06e498b3e560e7828a741d843f02a6b848c0b432af9ead18a57b03f
|
| 3 |
+
size 481
|
train_layer_index_28_mat_mlp.c_proj_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"total_flos":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 5.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 9.078073041438638e+17,
|
| 3 |
+
"train_loss": 1.6789688296530434,
|
| 4 |
+
"train_runtime": 12675.9905,
|
| 5 |
+
"train_samples": 294393,
|
| 6 |
+
"train_samples_per_second": 23.224,
|
| 7 |
+
"train_steps_per_second": 5.806
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -10,1188 +10,1188 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.006793616717732019,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 1.
|
| 16 |
-
"mean_token_accuracy": 0.
|
| 17 |
"step": 500
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.013587233435464039,
|
| 21 |
-
"grad_norm": 0.
|
| 22 |
-
"learning_rate": 0.
|
| 23 |
-
"loss": 1.
|
| 24 |
-
"mean_token_accuracy": 0.
|
| 25 |
"step": 1000
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 0.020380850153196056,
|
| 29 |
-
"grad_norm": 0.
|
| 30 |
-
"learning_rate": 0.
|
| 31 |
-
"loss": 1.
|
| 32 |
-
"mean_token_accuracy": 0.
|
| 33 |
"step": 1500
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.027174466870928077,
|
| 37 |
-
"grad_norm": 0.
|
| 38 |
-
"learning_rate": 0.
|
| 39 |
-
"loss": 1.
|
| 40 |
-
"mean_token_accuracy": 0.
|
| 41 |
"step": 2000
|
| 42 |
},
|
| 43 |
{
|
| 44 |
"epoch": 0.033968083588660096,
|
| 45 |
-
"grad_norm": 0.
|
| 46 |
"learning_rate": 0.0004830430174733009,
|
| 47 |
-
"loss": 1.
|
| 48 |
-
"mean_token_accuracy": 0.
|
| 49 |
"step": 2500
|
| 50 |
},
|
| 51 |
{
|
| 52 |
"epoch": 0.04076170030639211,
|
| 53 |
-
"grad_norm": 0.
|
| 54 |
-
"learning_rate": 0.
|
| 55 |
-
"loss": 1.
|
| 56 |
-
"mean_token_accuracy": 0.
|
| 57 |
"step": 3000
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.04755531702412413,
|
| 61 |
-
"grad_norm": 0.
|
| 62 |
-
"learning_rate": 0.
|
| 63 |
-
"loss": 1.
|
| 64 |
-
"mean_token_accuracy": 0.
|
| 65 |
"step": 3500
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.054348933741856155,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 1.
|
| 72 |
-
"mean_token_accuracy": 0.
|
| 73 |
"step": 4000
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.06114255045958817,
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
-
"learning_rate": 0.
|
| 79 |
-
"loss": 1.
|
| 80 |
-
"mean_token_accuracy": 0.
|
| 81 |
"step": 4500
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.06793616717732019,
|
| 85 |
-
"grad_norm": 0.
|
| 86 |
"learning_rate": 0.00046607244762085926,
|
| 87 |
-
"loss": 1.
|
| 88 |
-
"mean_token_accuracy": 0.
|
| 89 |
"step": 5000
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 0.07472978389505221,
|
| 93 |
-
"grad_norm": 0.
|
| 94 |
-
"learning_rate": 0.
|
| 95 |
-
"loss": 1.
|
| 96 |
-
"mean_token_accuracy": 0.
|
| 97 |
"step": 5500
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.08152340061278422,
|
| 101 |
-
"grad_norm": 0.
|
| 102 |
-
"learning_rate": 0.
|
| 103 |
-
"loss": 1.
|
| 104 |
-
"mean_token_accuracy": 0.
|
| 105 |
"step": 6000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.08831701733051625,
|
| 109 |
-
"grad_norm": 0.
|
| 110 |
-
"learning_rate": 0.
|
| 111 |
-
"loss": 1.
|
| 112 |
-
"mean_token_accuracy": 0.
|
| 113 |
"step": 6500
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 0.09511063404824827,
|
| 117 |
-
"grad_norm": 0.
|
| 118 |
-
"learning_rate": 0.
|
| 119 |
-
"loss": 1.
|
| 120 |
-
"mean_token_accuracy": 0.
|
| 121 |
"step": 7000
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.10190425076598028,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 1.
|
| 128 |
-
"mean_token_accuracy": 0.
|
| 129 |
"step": 7500
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.10869786748371231,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
-
"learning_rate": 0.
|
| 135 |
-
"loss": 1.
|
| 136 |
-
"mean_token_accuracy": 0.
|
| 137 |
"step": 8000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.11549148420144432,
|
| 141 |
-
"grad_norm": 0.
|
| 142 |
-
"learning_rate": 0.
|
| 143 |
-
"loss": 1.
|
| 144 |
-
"mean_token_accuracy": 0.
|
| 145 |
"step": 8500
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.12228510091917634,
|
| 149 |
-
"grad_norm": 0.
|
| 150 |
-
"learning_rate": 0.
|
| 151 |
-
"loss": 1.
|
| 152 |
-
"mean_token_accuracy": 0.
|
| 153 |
"step": 9000
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.12907871763690837,
|
| 157 |
-
"grad_norm": 0.
|
| 158 |
-
"learning_rate": 0.
|
| 159 |
-
"loss": 1.
|
| 160 |
-
"mean_token_accuracy": 0.
|
| 161 |
"step": 9500
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.13587233435464038,
|
| 165 |
-
"grad_norm": 0.
|
| 166 |
-
"learning_rate": 0.
|
| 167 |
-
"loss": 1.
|
| 168 |
-
"mean_token_accuracy": 0.
|
| 169 |
"step": 10000
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 0.1426659510723724,
|
| 173 |
-
"grad_norm": 0.
|
| 174 |
-
"learning_rate": 0.
|
| 175 |
-
"loss": 1.
|
| 176 |
-
"mean_token_accuracy": 0.
|
| 177 |
"step": 10500
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.14945956779010441,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss": 1.
|
| 184 |
-
"mean_token_accuracy": 0.
|
| 185 |
"step": 11000
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15625318450783643,
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
-
"learning_rate": 0.
|
| 191 |
-
"loss": 1.
|
| 192 |
-
"mean_token_accuracy": 0.
|
| 193 |
"step": 11500
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 0.16304680122556844,
|
| 197 |
-
"grad_norm": 0.
|
| 198 |
-
"learning_rate": 0.
|
| 199 |
-
"loss": 1.
|
| 200 |
-
"mean_token_accuracy": 0.
|
| 201 |
"step": 12000
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 0.1698404179433005,
|
| 205 |
-
"grad_norm": 0.
|
| 206 |
-
"learning_rate": 0.
|
| 207 |
-
"loss": 1.
|
| 208 |
-
"mean_token_accuracy": 0.
|
| 209 |
"step": 12500
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.1766340346610325,
|
| 213 |
-
"grad_norm": 0.
|
| 214 |
-
"learning_rate": 0.
|
| 215 |
-
"loss": 1.
|
| 216 |
-
"mean_token_accuracy": 0.
|
| 217 |
"step": 13000
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 0.18342765137876452,
|
| 221 |
-
"grad_norm": 0.
|
| 222 |
-
"learning_rate": 0.
|
| 223 |
-
"loss": 1.
|
| 224 |
-
"mean_token_accuracy": 0.
|
| 225 |
"step": 13500
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 0.19022126809649653,
|
| 229 |
-
"grad_norm": 0.
|
| 230 |
-
"learning_rate": 0.
|
| 231 |
-
"loss": 1.
|
| 232 |
-
"mean_token_accuracy": 0.
|
| 233 |
"step": 14000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.19701488481422855,
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss": 1.
|
| 240 |
-
"mean_token_accuracy": 0.
|
| 241 |
"step": 14500
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.20380850153196056,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
-
"learning_rate": 0.
|
| 247 |
-
"loss": 1.
|
| 248 |
-
"mean_token_accuracy": 0.
|
| 249 |
"step": 15000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 0.21060211824969258,
|
| 253 |
-
"grad_norm": 0.
|
| 254 |
-
"learning_rate": 0.
|
| 255 |
-
"loss": 1.
|
| 256 |
-
"mean_token_accuracy": 0.
|
| 257 |
"step": 15500
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.21739573496742462,
|
| 261 |
-
"grad_norm": 0.
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss": 1.
|
| 264 |
-
"mean_token_accuracy": 0.
|
| 265 |
"step": 16000
|
| 266 |
},
|
| 267 |
{
|
| 268 |
"epoch": 0.22418935168515663,
|
| 269 |
-
"grad_norm": 0.
|
| 270 |
-
"learning_rate": 0.
|
| 271 |
-
"loss": 1.
|
| 272 |
-
"mean_token_accuracy": 0.
|
| 273 |
"step": 16500
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 0.23098296840288865,
|
| 277 |
-
"grad_norm": 0.
|
| 278 |
-
"learning_rate": 0.
|
| 279 |
-
"loss": 1.
|
| 280 |
-
"mean_token_accuracy": 0.
|
| 281 |
"step": 17000
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 0.23777658512062066,
|
| 285 |
-
"grad_norm": 0.
|
| 286 |
-
"learning_rate": 0.
|
| 287 |
-
"loss": 1.
|
| 288 |
-
"mean_token_accuracy": 0.
|
| 289 |
"step": 17500
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.24457020183835268,
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate": 0.
|
| 295 |
-
"loss": 1.
|
| 296 |
-
"mean_token_accuracy": 0.
|
| 297 |
"step": 18000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.2513638185560847,
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
-
"learning_rate": 0.
|
| 303 |
-
"loss": 1.
|
| 304 |
-
"mean_token_accuracy": 0.
|
| 305 |
"step": 18500
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 0.25815743527381674,
|
| 309 |
-
"grad_norm": 0.
|
| 310 |
-
"learning_rate": 0.
|
| 311 |
-
"loss": 1.
|
| 312 |
-
"mean_token_accuracy": 0.
|
| 313 |
"step": 19000
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 0.2649510519915487,
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
-
"learning_rate": 0.
|
| 319 |
-
"loss": 1.
|
| 320 |
-
"mean_token_accuracy": 0.
|
| 321 |
"step": 19500
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 0.27174466870928077,
|
| 325 |
-
"grad_norm": 0.
|
| 326 |
-
"learning_rate": 0.
|
| 327 |
-
"loss": 1.
|
| 328 |
-
"mean_token_accuracy": 0.
|
| 329 |
"step": 20000
|
| 330 |
},
|
| 331 |
{
|
| 332 |
"epoch": 0.27853828542701276,
|
| 333 |
-
"grad_norm": 0.
|
| 334 |
-
"learning_rate": 0.
|
| 335 |
-
"loss": 1.
|
| 336 |
-
"mean_token_accuracy": 0.
|
| 337 |
"step": 20500
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 0.2853319021447448,
|
| 341 |
-
"grad_norm": 0.
|
| 342 |
-
"learning_rate": 0.
|
| 343 |
-
"loss": 1.
|
| 344 |
-
"mean_token_accuracy": 0.
|
| 345 |
"step": 21000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.29212551886247684,
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate": 0.
|
| 351 |
-
"loss": 1.
|
| 352 |
-
"mean_token_accuracy": 0.
|
| 353 |
"step": 21500
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.29891913558020883,
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
-
"learning_rate": 0.
|
| 359 |
-
"loss": 1.
|
| 360 |
-
"mean_token_accuracy": 0.
|
| 361 |
"step": 22000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.30571275229794087,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
-
"learning_rate": 0.
|
| 367 |
-
"loss": 1.
|
| 368 |
-
"mean_token_accuracy": 0.
|
| 369 |
"step": 22500
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 0.31250636901567286,
|
| 373 |
-
"grad_norm": 0.
|
| 374 |
-
"learning_rate": 0.
|
| 375 |
-
"loss": 1.
|
| 376 |
-
"mean_token_accuracy": 0.
|
| 377 |
"step": 23000
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 0.3192999857334049,
|
| 381 |
-
"grad_norm": 0.
|
| 382 |
-
"learning_rate": 0.
|
| 383 |
-
"loss": 1.
|
| 384 |
-
"mean_token_accuracy": 0.
|
| 385 |
"step": 23500
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 0.3260936024511369,
|
| 389 |
-
"grad_norm": 0.
|
| 390 |
-
"learning_rate": 0.
|
| 391 |
-
"loss": 1.
|
| 392 |
-
"mean_token_accuracy": 0.
|
| 393 |
"step": 24000
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 0.33288721916886893,
|
| 397 |
-
"grad_norm": 0.
|
| 398 |
-
"learning_rate": 0.
|
| 399 |
-
"loss": 1.
|
| 400 |
-
"mean_token_accuracy": 0.
|
| 401 |
"step": 24500
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.339680835886601,
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate": 0.
|
| 407 |
-
"loss": 1.
|
| 408 |
-
"mean_token_accuracy": 0.
|
| 409 |
"step": 25000
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.34647445260433296,
|
| 413 |
-
"grad_norm": 0.
|
| 414 |
-
"learning_rate": 0.
|
| 415 |
-
"loss": 1.
|
| 416 |
-
"mean_token_accuracy": 0.
|
| 417 |
"step": 25500
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.353268069322065,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss": 1.
|
| 424 |
-
"mean_token_accuracy": 0.
|
| 425 |
"step": 26000
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 0.360061686039797,
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
-
"learning_rate": 0.
|
| 431 |
-
"loss": 1.
|
| 432 |
-
"mean_token_accuracy": 0.
|
| 433 |
"step": 26500
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 0.36685530275752903,
|
| 437 |
-
"grad_norm": 0.
|
| 438 |
-
"learning_rate": 0.
|
| 439 |
-
"loss": 1.
|
| 440 |
-
"mean_token_accuracy": 0.
|
| 441 |
"step": 27000
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 0.373648919475261,
|
| 445 |
-
"grad_norm": 0.
|
| 446 |
-
"learning_rate": 0.
|
| 447 |
-
"loss": 1.
|
| 448 |
-
"mean_token_accuracy": 0.
|
| 449 |
"step": 27500
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 0.38044253619299306,
|
| 453 |
-
"grad_norm": 0.
|
| 454 |
-
"learning_rate": 0.
|
| 455 |
-
"loss": 1.
|
| 456 |
-
"mean_token_accuracy": 0.
|
| 457 |
"step": 28000
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.3872361529107251,
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate": 0.
|
| 463 |
-
"loss": 1.
|
| 464 |
-
"mean_token_accuracy": 0.
|
| 465 |
"step": 28500
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.3940297696284571,
|
| 469 |
-
"grad_norm": 0.
|
| 470 |
-
"learning_rate": 0.
|
| 471 |
-
"loss": 1.
|
| 472 |
-
"mean_token_accuracy": 0.
|
| 473 |
"step": 29000
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.40082338634618914,
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
-
"learning_rate": 0.
|
| 479 |
-
"loss": 1.
|
| 480 |
-
"mean_token_accuracy": 0.
|
| 481 |
"step": 29500
|
| 482 |
},
|
| 483 |
{
|
| 484 |
"epoch": 0.4076170030639211,
|
| 485 |
-
"grad_norm": 0.
|
| 486 |
-
"learning_rate": 0.
|
| 487 |
-
"loss": 1.
|
| 488 |
-
"mean_token_accuracy": 0.
|
| 489 |
"step": 30000
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.41441061978165317,
|
| 493 |
-
"grad_norm": 0.
|
| 494 |
-
"learning_rate": 0.
|
| 495 |
-
"loss": 1.
|
| 496 |
-
"mean_token_accuracy": 0.
|
| 497 |
"step": 30500
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 0.42120423649938515,
|
| 501 |
-
"grad_norm": 0.
|
| 502 |
-
"learning_rate": 0.
|
| 503 |
-
"loss": 1.
|
| 504 |
-
"mean_token_accuracy": 0.
|
| 505 |
"step": 31000
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 0.4279978532171172,
|
| 509 |
-
"grad_norm": 0.
|
| 510 |
-
"learning_rate": 0.
|
| 511 |
-
"loss": 1.
|
| 512 |
-
"mean_token_accuracy": 0.
|
| 513 |
"step": 31500
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 0.43479146993484924,
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
-
"learning_rate": 0.
|
| 519 |
-
"loss": 1.
|
| 520 |
-
"mean_token_accuracy": 0.
|
| 521 |
"step": 32000
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 0.4415850866525812,
|
| 525 |
-
"grad_norm": 0.
|
| 526 |
-
"learning_rate": 0.
|
| 527 |
-
"loss": 1.
|
| 528 |
-
"mean_token_accuracy": 0.
|
| 529 |
"step": 32500
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.44837870337031327,
|
| 533 |
-
"grad_norm": 0.
|
| 534 |
-
"learning_rate": 0.
|
| 535 |
-
"loss": 1.
|
| 536 |
-
"mean_token_accuracy": 0.
|
| 537 |
"step": 33000
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 0.45517232008804526,
|
| 541 |
-
"grad_norm": 0.
|
| 542 |
-
"learning_rate": 0.
|
| 543 |
-
"loss": 1.
|
| 544 |
-
"mean_token_accuracy": 0.
|
| 545 |
"step": 33500
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 0.4619659368057773,
|
| 549 |
-
"grad_norm": 0.
|
| 550 |
-
"learning_rate": 0.
|
| 551 |
-
"loss": 1.
|
| 552 |
-
"mean_token_accuracy": 0.
|
| 553 |
"step": 34000
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 0.4687595535235093,
|
| 557 |
-
"grad_norm": 0.
|
| 558 |
-
"learning_rate": 0.
|
| 559 |
-
"loss": 1.
|
| 560 |
-
"mean_token_accuracy": 0.
|
| 561 |
"step": 34500
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 0.47555317024124133,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
-
"learning_rate": 0.
|
| 567 |
-
"loss": 1.
|
| 568 |
-
"mean_token_accuracy": 0.
|
| 569 |
"step": 35000
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 0.4823467869589734,
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate": 0.
|
| 575 |
-
"loss": 1.
|
| 576 |
-
"mean_token_accuracy": 0.
|
| 577 |
"step": 35500
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 0.48914040367670536,
|
| 581 |
-
"grad_norm": 0.
|
| 582 |
-
"learning_rate": 0.
|
| 583 |
-
"loss": 1.
|
| 584 |
-
"mean_token_accuracy": 0.
|
| 585 |
"step": 36000
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.4959340203944374,
|
| 589 |
-
"grad_norm": 0.
|
| 590 |
-
"learning_rate": 0.
|
| 591 |
-
"loss": 1.
|
| 592 |
-
"mean_token_accuracy": 0.
|
| 593 |
"step": 36500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 0.5027276371121694,
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
-
"learning_rate": 0.
|
| 599 |
-
"loss": 1.
|
| 600 |
-
"mean_token_accuracy": 0.
|
| 601 |
"step": 37000
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 0.5095212538299014,
|
| 605 |
-
"grad_norm": 0.
|
| 606 |
-
"learning_rate": 0.
|
| 607 |
-
"loss": 1.
|
| 608 |
-
"mean_token_accuracy": 0.
|
| 609 |
"step": 37500
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 0.5163148705476335,
|
| 613 |
-
"grad_norm": 0.
|
| 614 |
-
"learning_rate": 0.
|
| 615 |
-
"loss": 1.
|
| 616 |
-
"mean_token_accuracy": 0.
|
| 617 |
"step": 38000
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 0.5231084872653654,
|
| 621 |
-
"grad_norm": 0.
|
| 622 |
-
"learning_rate": 0.
|
| 623 |
-
"loss": 1.
|
| 624 |
-
"mean_token_accuracy": 0.
|
| 625 |
"step": 38500
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 0.5299021039830974,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate": 0.
|
| 631 |
-
"loss": 1.
|
| 632 |
-
"mean_token_accuracy": 0.
|
| 633 |
"step": 39000
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 0.5366957207008295,
|
| 637 |
-
"grad_norm": 0.
|
| 638 |
-
"learning_rate": 0.
|
| 639 |
-
"loss": 1.
|
| 640 |
-
"mean_token_accuracy": 0.
|
| 641 |
"step": 39500
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 0.5434893374185615,
|
| 645 |
-
"grad_norm": 0.
|
| 646 |
-
"learning_rate": 0.
|
| 647 |
-
"loss": 1.
|
| 648 |
-
"mean_token_accuracy": 0.
|
| 649 |
"step": 40000
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.5502829541362936,
|
| 653 |
-
"grad_norm": 0.
|
| 654 |
-
"learning_rate": 0.
|
| 655 |
-
"loss": 1.
|
| 656 |
-
"mean_token_accuracy": 0.
|
| 657 |
"step": 40500
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 0.5570765708540255,
|
| 661 |
-
"grad_norm": 0.
|
| 662 |
-
"learning_rate": 0.
|
| 663 |
-
"loss": 1.
|
| 664 |
-
"mean_token_accuracy": 0.
|
| 665 |
"step": 41000
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 0.5638701875717576,
|
| 669 |
-
"grad_norm": 0.
|
| 670 |
-
"learning_rate": 0.
|
| 671 |
-
"loss": 1.
|
| 672 |
-
"mean_token_accuracy": 0.
|
| 673 |
"step": 41500
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 0.5706638042894896,
|
| 677 |
-
"grad_norm": 0.
|
| 678 |
-
"learning_rate": 0.
|
| 679 |
-
"loss": 1.
|
| 680 |
-
"mean_token_accuracy": 0.
|
| 681 |
"step": 42000
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 0.5774574210072216,
|
| 685 |
-
"grad_norm": 0.
|
| 686 |
-
"learning_rate": 0.
|
| 687 |
-
"loss": 1.
|
| 688 |
-
"mean_token_accuracy": 0.
|
| 689 |
"step": 42500
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 0.5842510377249537,
|
| 693 |
-
"grad_norm": 0.
|
| 694 |
-
"learning_rate": 0.
|
| 695 |
-
"loss": 1.
|
| 696 |
-
"mean_token_accuracy": 0.
|
| 697 |
"step": 43000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 0.5910446544426856,
|
| 701 |
-
"grad_norm": 0.
|
| 702 |
-
"learning_rate": 0.
|
| 703 |
-
"loss": 1.
|
| 704 |
-
"mean_token_accuracy": 0.
|
| 705 |
"step": 43500
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.5978382711604177,
|
| 709 |
-
"grad_norm": 0.
|
| 710 |
-
"learning_rate": 0.
|
| 711 |
-
"loss": 1.
|
| 712 |
-
"mean_token_accuracy": 0.
|
| 713 |
"step": 44000
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 0.6046318878781497,
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
-
"learning_rate": 0.
|
| 719 |
-
"loss": 1.
|
| 720 |
-
"mean_token_accuracy": 0.
|
| 721 |
"step": 44500
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 0.6114255045958817,
|
| 725 |
-
"grad_norm": 0.
|
| 726 |
-
"learning_rate": 0.
|
| 727 |
-
"loss": 1.
|
| 728 |
-
"mean_token_accuracy": 0.
|
| 729 |
"step": 45000
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 0.6182191213136137,
|
| 733 |
-
"grad_norm": 0.
|
| 734 |
-
"learning_rate": 0.
|
| 735 |
-
"loss": 1.
|
| 736 |
-
"mean_token_accuracy": 0.
|
| 737 |
"step": 45500
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 0.6250127380313457,
|
| 741 |
-
"grad_norm": 0.
|
| 742 |
-
"learning_rate": 0.
|
| 743 |
-
"loss": 1.
|
| 744 |
-
"mean_token_accuracy": 0.
|
| 745 |
"step": 46000
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 0.6318063547490778,
|
| 749 |
-
"grad_norm": 0.
|
| 750 |
-
"learning_rate": 0.
|
| 751 |
-
"loss": 1.
|
| 752 |
-
"mean_token_accuracy": 0.
|
| 753 |
"step": 46500
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 0.6385999714668098,
|
| 757 |
-
"grad_norm": 0.
|
| 758 |
-
"learning_rate": 0.
|
| 759 |
-
"loss": 1.
|
| 760 |
-
"mean_token_accuracy": 0.
|
| 761 |
"step": 47000
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.6453935881845418,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
-
"learning_rate": 0.
|
| 767 |
-
"loss": 1.
|
| 768 |
-
"mean_token_accuracy": 0.
|
| 769 |
"step": 47500
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 0.6521872049022738,
|
| 773 |
-
"grad_norm": 0.
|
| 774 |
-
"learning_rate": 0.
|
| 775 |
-
"loss": 1.
|
| 776 |
-
"mean_token_accuracy": 0.
|
| 777 |
"step": 48000
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 0.6589808216200058,
|
| 781 |
-
"grad_norm": 0.
|
| 782 |
-
"learning_rate": 0.
|
| 783 |
-
"loss": 1.
|
| 784 |
-
"mean_token_accuracy": 0.
|
| 785 |
"step": 48500
|
| 786 |
},
|
| 787 |
{
|
| 788 |
"epoch": 0.6657744383377379,
|
| 789 |
-
"grad_norm": 0.
|
| 790 |
-
"learning_rate": 0.
|
| 791 |
-
"loss": 1.
|
| 792 |
-
"mean_token_accuracy": 0.
|
| 793 |
"step": 49000
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 0.6725680550554699,
|
| 797 |
-
"grad_norm": 0.
|
| 798 |
-
"learning_rate": 0.
|
| 799 |
-
"loss": 1.
|
| 800 |
-
"mean_token_accuracy": 0.
|
| 801 |
"step": 49500
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 0.679361671773202,
|
| 805 |
-
"grad_norm": 0.
|
| 806 |
-
"learning_rate": 0.
|
| 807 |
-
"loss": 1.
|
| 808 |
-
"mean_token_accuracy": 0.
|
| 809 |
"step": 50000
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 0.6861552884909339,
|
| 813 |
-
"grad_norm": 0.
|
| 814 |
-
"learning_rate": 0.
|
| 815 |
-
"loss": 1.
|
| 816 |
-
"mean_token_accuracy": 0.
|
| 817 |
"step": 50500
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.6929489052086659,
|
| 821 |
-
"grad_norm": 0.
|
| 822 |
-
"learning_rate": 0.
|
| 823 |
-
"loss": 1.
|
| 824 |
-
"mean_token_accuracy": 0.
|
| 825 |
"step": 51000
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 0.699742521926398,
|
| 829 |
-
"grad_norm": 0.
|
| 830 |
-
"learning_rate": 0.
|
| 831 |
-
"loss": 1.
|
| 832 |
-
"mean_token_accuracy": 0.
|
| 833 |
"step": 51500
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 0.70653613864413,
|
| 837 |
-
"grad_norm": 0.
|
| 838 |
-
"learning_rate": 0.
|
| 839 |
-
"loss": 1.
|
| 840 |
-
"mean_token_accuracy": 0.
|
| 841 |
"step": 52000
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 0.7133297553618619,
|
| 845 |
-
"grad_norm": 0.
|
| 846 |
-
"learning_rate": 0.
|
| 847 |
-
"loss": 1.
|
| 848 |
-
"mean_token_accuracy": 0.
|
| 849 |
"step": 52500
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.720123372079594,
|
| 853 |
-
"grad_norm": 0.
|
| 854 |
-
"learning_rate": 0.
|
| 855 |
-
"loss": 1.
|
| 856 |
-
"mean_token_accuracy": 0.
|
| 857 |
"step": 53000
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 0.726916988797326,
|
| 861 |
-
"grad_norm": 0.
|
| 862 |
-
"learning_rate": 0.
|
| 863 |
-
"loss": 1.
|
| 864 |
-
"mean_token_accuracy": 0.
|
| 865 |
"step": 53500
|
| 866 |
},
|
| 867 |
{
|
| 868 |
"epoch": 0.7337106055150581,
|
| 869 |
-
"grad_norm": 0.
|
| 870 |
-
"learning_rate": 0.
|
| 871 |
-
"loss": 1.
|
| 872 |
-
"mean_token_accuracy": 0.
|
| 873 |
"step": 54000
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.7405042222327901,
|
| 877 |
-
"grad_norm": 0.
|
| 878 |
-
"learning_rate": 0.
|
| 879 |
-
"loss": 1.
|
| 880 |
-
"mean_token_accuracy": 0.
|
| 881 |
"step": 54500
|
| 882 |
},
|
| 883 |
{
|
| 884 |
"epoch": 0.747297838950522,
|
| 885 |
-
"grad_norm": 0.
|
| 886 |
-
"learning_rate": 0.
|
| 887 |
-
"loss": 1.
|
| 888 |
-
"mean_token_accuracy": 0.
|
| 889 |
"step": 55000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 0.7540914556682541,
|
| 893 |
-
"grad_norm": 0.
|
| 894 |
-
"learning_rate": 0.
|
| 895 |
-
"loss": 1.
|
| 896 |
-
"mean_token_accuracy": 0.
|
| 897 |
"step": 55500
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 0.7608850723859861,
|
| 901 |
-
"grad_norm": 0.
|
| 902 |
-
"learning_rate": 0.
|
| 903 |
-
"loss": 1.
|
| 904 |
-
"mean_token_accuracy": 0.
|
| 905 |
"step": 56000
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 0.7676786891037182,
|
| 909 |
-
"grad_norm": 0.
|
| 910 |
-
"learning_rate": 0.
|
| 911 |
-
"loss": 1.
|
| 912 |
-
"mean_token_accuracy": 0.
|
| 913 |
"step": 56500
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 0.7744723058214502,
|
| 917 |
-
"grad_norm": 0.
|
| 918 |
-
"learning_rate": 0.
|
| 919 |
-
"loss": 1.
|
| 920 |
-
"mean_token_accuracy": 0.
|
| 921 |
"step": 57000
|
| 922 |
},
|
| 923 |
{
|
| 924 |
"epoch": 0.7812659225391821,
|
| 925 |
-
"grad_norm": 0.
|
| 926 |
-
"learning_rate": 0.
|
| 927 |
-
"loss": 1.
|
| 928 |
-
"mean_token_accuracy": 0.
|
| 929 |
"step": 57500
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 0.7880595392569142,
|
| 933 |
-
"grad_norm": 0.
|
| 934 |
-
"learning_rate": 0.
|
| 935 |
-
"loss": 1.
|
| 936 |
-
"mean_token_accuracy": 0.
|
| 937 |
"step": 58000
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 0.7948531559746462,
|
| 941 |
-
"grad_norm": 0.
|
| 942 |
-
"learning_rate": 0.
|
| 943 |
-
"loss": 1.
|
| 944 |
-
"mean_token_accuracy": 0.
|
| 945 |
"step": 58500
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 0.8016467726923783,
|
| 949 |
-
"grad_norm": 0.
|
| 950 |
-
"learning_rate": 9.
|
| 951 |
-
"loss": 1.
|
| 952 |
-
"mean_token_accuracy": 0.
|
| 953 |
"step": 59000
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 0.8084403894101102,
|
| 957 |
-
"grad_norm": 0.
|
| 958 |
-
"learning_rate": 9.
|
| 959 |
-
"loss": 1.
|
| 960 |
-
"mean_token_accuracy": 0.
|
| 961 |
"step": 59500
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 0.8152340061278422,
|
| 965 |
-
"grad_norm": 0.
|
| 966 |
-
"learning_rate": 9.
|
| 967 |
-
"loss": 1.
|
| 968 |
-
"mean_token_accuracy": 0.
|
| 969 |
"step": 60000
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 0.8220276228455743,
|
| 973 |
-
"grad_norm": 0.
|
| 974 |
-
"learning_rate": 8.
|
| 975 |
-
"loss": 1.
|
| 976 |
-
"mean_token_accuracy": 0.
|
| 977 |
"step": 60500
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 0.8288212395633063,
|
| 981 |
-
"grad_norm": 0.
|
| 982 |
-
"learning_rate": 8.
|
| 983 |
-
"loss": 1.
|
| 984 |
-
"mean_token_accuracy": 0.
|
| 985 |
"step": 61000
|
| 986 |
},
|
| 987 |
{
|
| 988 |
"epoch": 0.8356148562810384,
|
| 989 |
-
"grad_norm": 0.
|
| 990 |
-
"learning_rate": 8.
|
| 991 |
-
"loss": 1.
|
| 992 |
-
"mean_token_accuracy": 0.
|
| 993 |
"step": 61500
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 0.8424084729987703,
|
| 997 |
-
"grad_norm": 0.
|
| 998 |
-
"learning_rate": 7.
|
| 999 |
-
"loss": 1.
|
| 1000 |
-
"mean_token_accuracy": 0.
|
| 1001 |
"step": 62000
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
"epoch": 0.8492020897165024,
|
| 1005 |
-
"grad_norm": 0.
|
| 1006 |
-
"learning_rate": 7.
|
| 1007 |
-
"loss": 1.
|
| 1008 |
-
"mean_token_accuracy": 0.
|
| 1009 |
"step": 62500
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 0.8559957064342344,
|
| 1013 |
-
"grad_norm": 0.
|
| 1014 |
-
"learning_rate": 7.
|
| 1015 |
-
"loss": 1.
|
| 1016 |
-
"mean_token_accuracy": 0.
|
| 1017 |
"step": 63000
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
"epoch": 0.8627893231519664,
|
| 1021 |
-
"grad_norm": 0.
|
| 1022 |
-
"learning_rate": 6.
|
| 1023 |
-
"loss": 1.
|
| 1024 |
-
"mean_token_accuracy": 0.
|
| 1025 |
"step": 63500
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
"epoch": 0.8695829398696985,
|
| 1029 |
-
"grad_norm": 0.
|
| 1030 |
-
"learning_rate": 6.
|
| 1031 |
-
"loss": 1.
|
| 1032 |
-
"mean_token_accuracy": 0.
|
| 1033 |
"step": 64000
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 0.8763765565874304,
|
| 1037 |
-
"grad_norm": 0.
|
| 1038 |
-
"learning_rate": 6.
|
| 1039 |
-
"loss": 1.
|
| 1040 |
-
"mean_token_accuracy": 0.
|
| 1041 |
"step": 64500
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 0.8831701733051625,
|
| 1045 |
-
"grad_norm": 0.
|
| 1046 |
-
"learning_rate": 5.
|
| 1047 |
-
"loss": 1.
|
| 1048 |
-
"mean_token_accuracy": 0.
|
| 1049 |
"step": 65000
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 0.8899637900228945,
|
| 1053 |
-
"grad_norm": 0.
|
| 1054 |
-
"learning_rate": 5.
|
| 1055 |
-
"loss": 1.
|
| 1056 |
-
"mean_token_accuracy": 0.
|
| 1057 |
"step": 65500
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 0.8967574067406265,
|
| 1061 |
-
"grad_norm": 0.
|
| 1062 |
-
"learning_rate": 5.
|
| 1063 |
-
"loss": 1.
|
| 1064 |
-
"mean_token_accuracy": 0.
|
| 1065 |
"step": 66000
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 0.9035510234583585,
|
| 1069 |
-
"grad_norm": 0.
|
| 1070 |
-
"learning_rate": 4.
|
| 1071 |
-
"loss": 1.
|
| 1072 |
-
"mean_token_accuracy": 0.
|
| 1073 |
"step": 66500
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 0.9103446401760905,
|
| 1077 |
-
"grad_norm": 0.
|
| 1078 |
-
"learning_rate": 4.
|
| 1079 |
-
"loss": 1.
|
| 1080 |
-
"mean_token_accuracy": 0.
|
| 1081 |
"step": 67000
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
"epoch": 0.9171382568938226,
|
| 1085 |
-
"grad_norm": 0.
|
| 1086 |
-
"learning_rate": 4.
|
| 1087 |
-
"loss": 1.
|
| 1088 |
-
"mean_token_accuracy": 0.
|
| 1089 |
"step": 67500
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
"epoch": 0.9239318736115546,
|
| 1093 |
-
"grad_norm": 0.
|
| 1094 |
-
"learning_rate": 3.
|
| 1095 |
-
"loss": 1.
|
| 1096 |
-
"mean_token_accuracy": 0.
|
| 1097 |
"step": 68000
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
"epoch": 0.9307254903292866,
|
| 1101 |
-
"grad_norm": 0.
|
| 1102 |
-
"learning_rate": 3.
|
| 1103 |
-
"loss": 1.
|
| 1104 |
-
"mean_token_accuracy": 0.
|
| 1105 |
"step": 68500
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 0.9375191070470186,
|
| 1109 |
-
"grad_norm": 0.
|
| 1110 |
-
"learning_rate": 3.
|
| 1111 |
-
"loss": 1.
|
| 1112 |
-
"mean_token_accuracy": 0.
|
| 1113 |
"step": 69000
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 0.9443127237647506,
|
| 1117 |
-
"grad_norm": 0.
|
| 1118 |
-
"learning_rate": 2.
|
| 1119 |
-
"loss": 1.
|
| 1120 |
-
"mean_token_accuracy": 0.
|
| 1121 |
"step": 69500
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 0.9511063404824827,
|
| 1125 |
-
"grad_norm": 0.
|
| 1126 |
-
"learning_rate": 2.
|
| 1127 |
-
"loss": 1.
|
| 1128 |
-
"mean_token_accuracy": 0.
|
| 1129 |
"step": 70000
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 0.9578999572002147,
|
| 1133 |
-
"grad_norm": 0.
|
| 1134 |
-
"learning_rate": 2.
|
| 1135 |
-
"loss": 1.
|
| 1136 |
-
"mean_token_accuracy": 0.
|
| 1137 |
"step": 70500
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 0.9646935739179467,
|
| 1141 |
-
"grad_norm": 0.
|
| 1142 |
-
"learning_rate": 1.
|
| 1143 |
-
"loss": 1.
|
| 1144 |
-
"mean_token_accuracy": 0.
|
| 1145 |
"step": 71000
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 0.9714871906356787,
|
| 1149 |
-
"grad_norm": 0.
|
| 1150 |
-
"learning_rate": 1.
|
| 1151 |
-
"loss": 1.
|
| 1152 |
-
"mean_token_accuracy": 0.
|
| 1153 |
"step": 71500
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
"epoch": 0.9782808073534107,
|
| 1157 |
-
"grad_norm": 0.
|
| 1158 |
-
"learning_rate": 1.
|
| 1159 |
-
"loss": 1.
|
| 1160 |
-
"mean_token_accuracy": 0.
|
| 1161 |
"step": 72000
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
"epoch": 0.9850744240711428,
|
| 1165 |
-
"grad_norm": 0.
|
| 1166 |
-
"learning_rate": 8.
|
| 1167 |
-
"loss": 1.
|
| 1168 |
-
"mean_token_accuracy": 0.
|
| 1169 |
"step": 72500
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
"epoch": 0.9918680407888748,
|
| 1173 |
-
"grad_norm": 0.
|
| 1174 |
-
"learning_rate":
|
| 1175 |
-
"loss": 1.
|
| 1176 |
-
"mean_token_accuracy": 0.
|
| 1177 |
"step": 73000
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 0.9986616575066067,
|
| 1181 |
-
"grad_norm": 0.
|
| 1182 |
-
"learning_rate": 1.
|
| 1183 |
-
"loss": 1.
|
| 1184 |
-
"mean_token_accuracy": 0.
|
| 1185 |
"step": 73500
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 0.9999932063832823,
|
| 1189 |
-
"mean_token_accuracy": 0.
|
| 1190 |
"step": 73598,
|
| 1191 |
-
"total_flos": 9.
|
| 1192 |
-
"train_loss": 1.
|
| 1193 |
-
"train_runtime":
|
| 1194 |
-
"train_samples_per_second": 23.
|
| 1195 |
"train_steps_per_second": 5.806
|
| 1196 |
}
|
| 1197 |
],
|
|
@@ -1212,7 +1212,7 @@
|
|
| 1212 |
"attributes": {}
|
| 1213 |
}
|
| 1214 |
},
|
| 1215 |
-
"total_flos": 9.
|
| 1216 |
"train_batch_size": 1,
|
| 1217 |
"trial_name": null,
|
| 1218 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.006793616717732019,
|
| 13 |
+
"grad_norm": 0.4056946635246277,
|
| 14 |
+
"learning_rate": 0.000496623549552977,
|
| 15 |
+
"loss": 1.7688,
|
| 16 |
+
"mean_token_accuracy": 0.5910043245553971,
|
| 17 |
"step": 500
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.013587233435464039,
|
| 21 |
+
"grad_norm": 0.4064154028892517,
|
| 22 |
+
"learning_rate": 0.0004932267181173402,
|
| 23 |
+
"loss": 1.7339,
|
| 24 |
+
"mean_token_accuracy": 0.59465124219656,
|
| 25 |
"step": 1000
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 0.020380850153196056,
|
| 29 |
+
"grad_norm": 0.42659252882003784,
|
| 30 |
+
"learning_rate": 0.0004898298866817033,
|
| 31 |
+
"loss": 1.7231,
|
| 32 |
+
"mean_token_accuracy": 0.5960495759248734,
|
| 33 |
"step": 1500
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.027174466870928077,
|
| 37 |
+
"grad_norm": 0.37063390016555786,
|
| 38 |
+
"learning_rate": 0.00048643305524606644,
|
| 39 |
+
"loss": 1.7228,
|
| 40 |
+
"mean_token_accuracy": 0.5968798612058163,
|
| 41 |
"step": 2000
|
| 42 |
},
|
| 43 |
{
|
| 44 |
"epoch": 0.033968083588660096,
|
| 45 |
+
"grad_norm": 0.4430277943611145,
|
| 46 |
"learning_rate": 0.0004830430174733009,
|
| 47 |
+
"loss": 1.7177,
|
| 48 |
+
"mean_token_accuracy": 0.5959446367025375,
|
| 49 |
"step": 2500
|
| 50 |
},
|
| 51 |
{
|
| 52 |
"epoch": 0.04076170030639211,
|
| 53 |
+
"grad_norm": 0.5757402181625366,
|
| 54 |
+
"learning_rate": 0.00047965297970053533,
|
| 55 |
+
"loss": 1.7291,
|
| 56 |
+
"mean_token_accuracy": 0.5950138145983219,
|
| 57 |
"step": 3000
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.04755531702412413,
|
| 61 |
+
"grad_norm": 0.465655654668808,
|
| 62 |
+
"learning_rate": 0.0004762561482648985,
|
| 63 |
+
"loss": 1.7146,
|
| 64 |
+
"mean_token_accuracy": 0.5979318037629128,
|
| 65 |
"step": 3500
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.054348933741856155,
|
| 69 |
+
"grad_norm": 0.3922131061553955,
|
| 70 |
+
"learning_rate": 0.00047285931682926167,
|
| 71 |
+
"loss": 1.7188,
|
| 72 |
+
"mean_token_accuracy": 0.593923301845789,
|
| 73 |
"step": 4000
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.06114255045958817,
|
| 77 |
+
"grad_norm": 0.5041131377220154,
|
| 78 |
+
"learning_rate": 0.00046946248539362484,
|
| 79 |
+
"loss": 1.6899,
|
| 80 |
+
"mean_token_accuracy": 0.6019016440212727,
|
| 81 |
"step": 4500
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.06793616717732019,
|
| 85 |
+
"grad_norm": 0.3338882327079773,
|
| 86 |
"learning_rate": 0.00046607244762085926,
|
| 87 |
+
"loss": 1.7051,
|
| 88 |
+
"mean_token_accuracy": 0.5998869875967503,
|
| 89 |
"step": 5000
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 0.07472978389505221,
|
| 93 |
+
"grad_norm": 0.4194291830062866,
|
| 94 |
+
"learning_rate": 0.00046267561618522243,
|
| 95 |
+
"loss": 1.7102,
|
| 96 |
+
"mean_token_accuracy": 0.5985791925489903,
|
| 97 |
"step": 5500
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.08152340061278422,
|
| 101 |
+
"grad_norm": 0.42370814085006714,
|
| 102 |
+
"learning_rate": 0.0004592991657381994,
|
| 103 |
+
"loss": 1.7104,
|
| 104 |
+
"mean_token_accuracy": 0.5979952279627323,
|
| 105 |
"step": 6000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.08831701733051625,
|
| 109 |
+
"grad_norm": 0.43259018659591675,
|
| 110 |
+
"learning_rate": 0.0004559023343025626,
|
| 111 |
+
"loss": 1.7007,
|
| 112 |
+
"mean_token_accuracy": 0.6013973373174667,
|
| 113 |
"step": 6500
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 0.09511063404824827,
|
| 117 |
+
"grad_norm": 0.4605717957019806,
|
| 118 |
+
"learning_rate": 0.00045250550286692574,
|
| 119 |
+
"loss": 1.7022,
|
| 120 |
+
"mean_token_accuracy": 0.5996414051353931,
|
| 121 |
"step": 7000
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.10190425076598028,
|
| 125 |
+
"grad_norm": 0.4562588632106781,
|
| 126 |
+
"learning_rate": 0.0004491086714312889,
|
| 127 |
+
"loss": 1.7047,
|
| 128 |
+
"mean_token_accuracy": 0.5984801447093486,
|
| 129 |
"step": 7500
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.10869786748371231,
|
| 133 |
+
"grad_norm": 0.43784981966018677,
|
| 134 |
+
"learning_rate": 0.0004457254273213946,
|
| 135 |
+
"loss": 1.6963,
|
| 136 |
+
"mean_token_accuracy": 0.6002669661641121,
|
| 137 |
"step": 8000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.11549148420144432,
|
| 141 |
+
"grad_norm": 0.4280397891998291,
|
| 142 |
+
"learning_rate": 0.0004423285958857578,
|
| 143 |
+
"loss": 1.7075,
|
| 144 |
+
"mean_token_accuracy": 0.5975034932494163,
|
| 145 |
"step": 8500
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.12228510091917634,
|
| 149 |
+
"grad_norm": 0.4308539628982544,
|
| 150 |
+
"learning_rate": 0.000438931764450121,
|
| 151 |
+
"loss": 1.6968,
|
| 152 |
+
"mean_token_accuracy": 0.599761339455843,
|
| 153 |
"step": 9000
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.12907871763690837,
|
| 157 |
+
"grad_norm": 0.37971165776252747,
|
| 158 |
+
"learning_rate": 0.0004355349330144841,
|
| 159 |
+
"loss": 1.6893,
|
| 160 |
+
"mean_token_accuracy": 0.602813461214304,
|
| 161 |
"step": 9500
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.13587233435464038,
|
| 165 |
+
"grad_norm": 0.42085403203964233,
|
| 166 |
+
"learning_rate": 0.0004321516889045898,
|
| 167 |
+
"loss": 1.6944,
|
| 168 |
+
"mean_token_accuracy": 0.5994773605763912,
|
| 169 |
"step": 10000
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 0.1426659510723724,
|
| 173 |
+
"grad_norm": 0.5180936455726624,
|
| 174 |
+
"learning_rate": 0.00042876165113182424,
|
| 175 |
+
"loss": 1.6939,
|
| 176 |
+
"mean_token_accuracy": 0.5987960757911205,
|
| 177 |
"step": 10500
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.14945956779010441,
|
| 181 |
+
"grad_norm": 0.49184420704841614,
|
| 182 |
+
"learning_rate": 0.0004253648196961874,
|
| 183 |
+
"loss": 1.6819,
|
| 184 |
+
"mean_token_accuracy": 0.6046352021694184,
|
| 185 |
"step": 11000
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15625318450783643,
|
| 189 |
+
"grad_norm": 0.4975549578666687,
|
| 190 |
+
"learning_rate": 0.0004219679882605506,
|
| 191 |
+
"loss": 1.6899,
|
| 192 |
+
"mean_token_accuracy": 0.6011166540384293,
|
| 193 |
"step": 11500
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 0.16304680122556844,
|
| 197 |
+
"grad_norm": 0.583638072013855,
|
| 198 |
+
"learning_rate": 0.00041857115682491375,
|
| 199 |
+
"loss": 1.6809,
|
| 200 |
+
"mean_token_accuracy": 0.6029804965853691,
|
| 201 |
"step": 12000
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 0.1698404179433005,
|
| 205 |
+
"grad_norm": 0.46068498492240906,
|
| 206 |
+
"learning_rate": 0.0004151947063778907,
|
| 207 |
+
"loss": 1.6924,
|
| 208 |
+
"mean_token_accuracy": 0.5984785186350345,
|
| 209 |
"step": 12500
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.1766340346610325,
|
| 213 |
+
"grad_norm": 0.4548553228378296,
|
| 214 |
+
"learning_rate": 0.00041179787494225384,
|
| 215 |
+
"loss": 1.7141,
|
| 216 |
+
"mean_token_accuracy": 0.5960147912800312,
|
| 217 |
"step": 13000
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 0.18342765137876452,
|
| 221 |
+
"grad_norm": 0.35259950160980225,
|
| 222 |
+
"learning_rate": 0.0004084078371694883,
|
| 223 |
+
"loss": 1.7108,
|
| 224 |
+
"mean_token_accuracy": 0.59765238276124,
|
| 225 |
"step": 13500
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 0.19022126809649653,
|
| 229 |
+
"grad_norm": 0.550769567489624,
|
| 230 |
+
"learning_rate": 0.0004050110057338515,
|
| 231 |
+
"loss": 1.6857,
|
| 232 |
+
"mean_token_accuracy": 0.6022021851241589,
|
| 233 |
"step": 14000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.19701488481422855,
|
| 237 |
+
"grad_norm": 0.46585261821746826,
|
| 238 |
+
"learning_rate": 0.0004016141742982146,
|
| 239 |
+
"loss": 1.68,
|
| 240 |
+
"mean_token_accuracy": 0.6016397434175015,
|
| 241 |
"step": 14500
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.20380850153196056,
|
| 245 |
+
"grad_norm": 0.463501513004303,
|
| 246 |
+
"learning_rate": 0.00039821734286257777,
|
| 247 |
+
"loss": 1.6799,
|
| 248 |
+
"mean_token_accuracy": 0.6029948250949383,
|
| 249 |
"step": 15000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 0.21060211824969258,
|
| 253 |
+
"grad_norm": 0.4906240701675415,
|
| 254 |
+
"learning_rate": 0.000394820511426941,
|
| 255 |
+
"loss": 1.699,
|
| 256 |
+
"mean_token_accuracy": 0.6011771605908871,
|
| 257 |
"step": 15500
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.21739573496742462,
|
| 261 |
+
"grad_norm": 0.49514085054397583,
|
| 262 |
+
"learning_rate": 0.0003914236799913041,
|
| 263 |
+
"loss": 1.6805,
|
| 264 |
+
"mean_token_accuracy": 0.6031656765639782,
|
| 265 |
"step": 16000
|
| 266 |
},
|
| 267 |
{
|
| 268 |
"epoch": 0.22418935168515663,
|
| 269 |
+
"grad_norm": 0.5545573830604553,
|
| 270 |
+
"learning_rate": 0.00038804043588140983,
|
| 271 |
+
"loss": 1.6914,
|
| 272 |
+
"mean_token_accuracy": 0.6014098982512951,
|
| 273 |
"step": 16500
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 0.23098296840288865,
|
| 277 |
+
"grad_norm": 0.4257826805114746,
|
| 278 |
+
"learning_rate": 0.00038465039810864425,
|
| 279 |
+
"loss": 1.6914,
|
| 280 |
+
"mean_token_accuracy": 0.6002418109178543,
|
| 281 |
"step": 17000
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 0.23777658512062066,
|
| 285 |
+
"grad_norm": 0.47851425409317017,
|
| 286 |
+
"learning_rate": 0.0003812535666730074,
|
| 287 |
+
"loss": 1.6871,
|
| 288 |
+
"mean_token_accuracy": 0.6003048142194748,
|
| 289 |
"step": 17500
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.24457020183835268,
|
| 293 |
+
"grad_norm": 0.438859760761261,
|
| 294 |
+
"learning_rate": 0.0003778567352373706,
|
| 295 |
+
"loss": 1.6891,
|
| 296 |
+
"mean_token_accuracy": 0.5999817061126232,
|
| 297 |
"step": 18000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.2513638185560847,
|
| 301 |
+
"grad_norm": 0.39294207096099854,
|
| 302 |
+
"learning_rate": 0.0003744599038017337,
|
| 303 |
+
"loss": 1.6881,
|
| 304 |
+
"mean_token_accuracy": 0.6006654644608498,
|
| 305 |
"step": 18500
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 0.25815743527381674,
|
| 309 |
+
"grad_norm": 0.40909263491630554,
|
| 310 |
+
"learning_rate": 0.0003710766596918395,
|
| 311 |
+
"loss": 1.6846,
|
| 312 |
+
"mean_token_accuracy": 0.6008345121741295,
|
| 313 |
"step": 19000
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 0.2649510519915487,
|
| 317 |
+
"grad_norm": 0.48700499534606934,
|
| 318 |
+
"learning_rate": 0.0003676798282562026,
|
| 319 |
+
"loss": 1.6855,
|
| 320 |
+
"mean_token_accuracy": 0.6010488954186439,
|
| 321 |
"step": 19500
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 0.27174466870928077,
|
| 325 |
+
"grad_norm": 0.5421476364135742,
|
| 326 |
+
"learning_rate": 0.0003642965841463083,
|
| 327 |
+
"loss": 1.6855,
|
| 328 |
+
"mean_token_accuracy": 0.599077152788639,
|
| 329 |
"step": 20000
|
| 330 |
},
|
| 331 |
{
|
| 332 |
"epoch": 0.27853828542701276,
|
| 333 |
+
"grad_norm": 0.7285926342010498,
|
| 334 |
+
"learning_rate": 0.0003608997527106715,
|
| 335 |
+
"loss": 1.6952,
|
| 336 |
+
"mean_token_accuracy": 0.5991786162257194,
|
| 337 |
"step": 20500
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 0.2853319021447448,
|
| 341 |
+
"grad_norm": 0.4716951549053192,
|
| 342 |
+
"learning_rate": 0.0003575165086007772,
|
| 343 |
+
"loss": 1.6799,
|
| 344 |
+
"mean_token_accuracy": 0.6026499591767788,
|
| 345 |
"step": 21000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.29212551886247684,
|
| 349 |
+
"grad_norm": 0.4939189851284027,
|
| 350 |
+
"learning_rate": 0.0003541196771651404,
|
| 351 |
+
"loss": 1.6927,
|
| 352 |
+
"mean_token_accuracy": 0.5991746371984482,
|
| 353 |
"step": 21500
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.29891913558020883,
|
| 357 |
+
"grad_norm": 0.3434264361858368,
|
| 358 |
+
"learning_rate": 0.00035073643305524605,
|
| 359 |
+
"loss": 1.6868,
|
| 360 |
+
"mean_token_accuracy": 0.6017061140835285,
|
| 361 |
"step": 22000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.30571275229794087,
|
| 365 |
+
"grad_norm": 0.4739142060279846,
|
| 366 |
+
"learning_rate": 0.0003473396016196093,
|
| 367 |
+
"loss": 1.6943,
|
| 368 |
+
"mean_token_accuracy": 0.5989596297442913,
|
| 369 |
"step": 22500
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 0.31250636901567286,
|
| 373 |
+
"grad_norm": 0.6250849962234497,
|
| 374 |
+
"learning_rate": 0.00034395635750971495,
|
| 375 |
+
"loss": 1.6738,
|
| 376 |
+
"mean_token_accuracy": 0.6042404849231243,
|
| 377 |
"step": 23000
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 0.3192999857334049,
|
| 381 |
+
"grad_norm": 0.4180070459842682,
|
| 382 |
+
"learning_rate": 0.00034055952607407806,
|
| 383 |
+
"loss": 1.6758,
|
| 384 |
+
"mean_token_accuracy": 0.6041967616379261,
|
| 385 |
"step": 23500
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 0.3260936024511369,
|
| 389 |
+
"grad_norm": 0.4894815981388092,
|
| 390 |
+
"learning_rate": 0.00033717628196418384,
|
| 391 |
+
"loss": 1.6799,
|
| 392 |
+
"mean_token_accuracy": 0.6030513152778149,
|
| 393 |
"step": 24000
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 0.33288721916886893,
|
| 397 |
+
"grad_norm": 0.4394868314266205,
|
| 398 |
+
"learning_rate": 0.00033377945052854695,
|
| 399 |
+
"loss": 1.6896,
|
| 400 |
+
"mean_token_accuracy": 0.5995272571742535,
|
| 401 |
"step": 24500
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.339680835886601,
|
| 405 |
+
"grad_norm": 0.4560585021972656,
|
| 406 |
+
"learning_rate": 0.0003303962064186527,
|
| 407 |
+
"loss": 1.6761,
|
| 408 |
+
"mean_token_accuracy": 0.602882645457983,
|
| 409 |
"step": 25000
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 0.34647445260433296,
|
| 413 |
+
"grad_norm": 0.4372274875640869,
|
| 414 |
+
"learning_rate": 0.00032699937498301585,
|
| 415 |
+
"loss": 1.6729,
|
| 416 |
+
"mean_token_accuracy": 0.6026374279558658,
|
| 417 |
"step": 25500
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.353268069322065,
|
| 421 |
+
"grad_norm": 0.40903204679489136,
|
| 422 |
+
"learning_rate": 0.00032361613087312157,
|
| 423 |
+
"loss": 1.6732,
|
| 424 |
+
"mean_token_accuracy": 0.6029972824752331,
|
| 425 |
"step": 26000
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 0.360061686039797,
|
| 429 |
+
"grad_norm": 0.4154110550880432,
|
| 430 |
+
"learning_rate": 0.00032021929943748474,
|
| 431 |
+
"loss": 1.6889,
|
| 432 |
+
"mean_token_accuracy": 0.6004891120493412,
|
| 433 |
"step": 26500
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 0.36685530275752903,
|
| 437 |
+
"grad_norm": 0.47997039556503296,
|
| 438 |
+
"learning_rate": 0.0003168360553275904,
|
| 439 |
+
"loss": 1.6688,
|
| 440 |
+
"mean_token_accuracy": 0.6037946022450924,
|
| 441 |
"step": 27000
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 0.373648919475261,
|
| 445 |
+
"grad_norm": 0.4413166046142578,
|
| 446 |
+
"learning_rate": 0.00031343922389195363,
|
| 447 |
+
"loss": 1.6656,
|
| 448 |
+
"mean_token_accuracy": 0.6043658980131149,
|
| 449 |
"step": 27500
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 0.38044253619299306,
|
| 453 |
+
"grad_norm": 0.3751048147678375,
|
| 454 |
+
"learning_rate": 0.0003100559797820593,
|
| 455 |
+
"loss": 1.6813,
|
| 456 |
+
"mean_token_accuracy": 0.601385792285204,
|
| 457 |
"step": 28000
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.3872361529107251,
|
| 461 |
+
"grad_norm": 0.42594870924949646,
|
| 462 |
+
"learning_rate": 0.0003066591483464224,
|
| 463 |
+
"loss": 1.693,
|
| 464 |
+
"mean_token_accuracy": 0.601170395553112,
|
| 465 |
"step": 28500
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 0.3940297696284571,
|
| 469 |
+
"grad_norm": 0.5058168172836304,
|
| 470 |
+
"learning_rate": 0.0003032759042365282,
|
| 471 |
+
"loss": 1.6761,
|
| 472 |
+
"mean_token_accuracy": 0.6033368219733238,
|
| 473 |
"step": 29000
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.40082338634618914,
|
| 477 |
+
"grad_norm": 0.5077099800109863,
|
| 478 |
+
"learning_rate": 0.0002998790728008913,
|
| 479 |
+
"loss": 1.6726,
|
| 480 |
+
"mean_token_accuracy": 0.6028481238484382,
|
| 481 |
"step": 29500
|
| 482 |
},
|
| 483 |
{
|
| 484 |
"epoch": 0.4076170030639211,
|
| 485 |
+
"grad_norm": 0.4526306986808777,
|
| 486 |
+
"learning_rate": 0.00029649582869099704,
|
| 487 |
+
"loss": 1.6955,
|
| 488 |
+
"mean_token_accuracy": 0.6015688245892524,
|
| 489 |
"step": 30000
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 0.41441061978165317,
|
| 493 |
+
"grad_norm": 0.7025688886642456,
|
| 494 |
+
"learning_rate": 0.0002930989972553602,
|
| 495 |
+
"loss": 1.6804,
|
| 496 |
+
"mean_token_accuracy": 0.6013101308941841,
|
| 497 |
"step": 30500
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 0.42120423649938515,
|
| 501 |
+
"grad_norm": 0.37353208661079407,
|
| 502 |
+
"learning_rate": 0.0002897089594825946,
|
| 503 |
+
"loss": 1.6616,
|
| 504 |
+
"mean_token_accuracy": 0.6058516998887062,
|
| 505 |
"step": 31000
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 0.4279978532171172,
|
| 509 |
+
"grad_norm": 0.40674665570259094,
|
| 510 |
+
"learning_rate": 0.0002863121280469578,
|
| 511 |
+
"loss": 1.6714,
|
| 512 |
+
"mean_token_accuracy": 0.601816111266613,
|
| 513 |
"step": 31500
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 0.43479146993484924,
|
| 517 |
+
"grad_norm": 0.49832838773727417,
|
| 518 |
+
"learning_rate": 0.00028291529661132096,
|
| 519 |
+
"loss": 1.6941,
|
| 520 |
+
"mean_token_accuracy": 0.5991431129276752,
|
| 521 |
"step": 32000
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 0.4415850866525812,
|
| 525 |
+
"grad_norm": 0.47476768493652344,
|
| 526 |
+
"learning_rate": 0.00027951846517568414,
|
| 527 |
+
"loss": 1.6857,
|
| 528 |
+
"mean_token_accuracy": 0.5986996866464614,
|
| 529 |
"step": 32500
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.44837870337031327,
|
| 533 |
+
"grad_norm": 0.4772721230983734,
|
| 534 |
+
"learning_rate": 0.0002761352210657898,
|
| 535 |
+
"loss": 1.6699,
|
| 536 |
+
"mean_token_accuracy": 0.6043048515915871,
|
| 537 |
"step": 33000
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 0.45517232008804526,
|
| 541 |
+
"grad_norm": 0.43461740016937256,
|
| 542 |
+
"learning_rate": 0.00027273838963015303,
|
| 543 |
+
"loss": 1.6807,
|
| 544 |
+
"mean_token_accuracy": 0.6015589760541916,
|
| 545 |
"step": 33500
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 0.4619659368057773,
|
| 549 |
+
"grad_norm": 0.49901214241981506,
|
| 550 |
+
"learning_rate": 0.0002693551455202587,
|
| 551 |
+
"loss": 1.6693,
|
| 552 |
+
"mean_token_accuracy": 0.603463482439518,
|
| 553 |
"step": 34000
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 0.4687595535235093,
|
| 557 |
+
"grad_norm": 0.5628573298454285,
|
| 558 |
+
"learning_rate": 0.00026595831408462187,
|
| 559 |
+
"loss": 1.661,
|
| 560 |
+
"mean_token_accuracy": 0.6067080739736557,
|
| 561 |
"step": 34500
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 0.47555317024124133,
|
| 565 |
+
"grad_norm": 0.517264187335968,
|
| 566 |
+
"learning_rate": 0.0002625750699747276,
|
| 567 |
+
"loss": 1.6679,
|
| 568 |
+
"mean_token_accuracy": 0.6032546425163746,
|
| 569 |
"step": 35000
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 0.4823467869589734,
|
| 573 |
+
"grad_norm": 0.45866549015045166,
|
| 574 |
+
"learning_rate": 0.0002591782385390907,
|
| 575 |
+
"loss": 1.6614,
|
| 576 |
+
"mean_token_accuracy": 0.6066102555394173,
|
| 577 |
"step": 35500
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 0.48914040367670536,
|
| 581 |
+
"grad_norm": 0.4672979414463043,
|
| 582 |
+
"learning_rate": 0.0002557949944291965,
|
| 583 |
+
"loss": 1.6736,
|
| 584 |
+
"mean_token_accuracy": 0.6030917878448964,
|
| 585 |
"step": 36000
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.4959340203944374,
|
| 589 |
+
"grad_norm": 0.43743544816970825,
|
| 590 |
+
"learning_rate": 0.00025240495665643085,
|
| 591 |
+
"loss": 1.6652,
|
| 592 |
+
"mean_token_accuracy": 0.6046005863845348,
|
| 593 |
"step": 36500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 0.5027276371121694,
|
| 597 |
+
"grad_norm": 0.4721488058567047,
|
| 598 |
+
"learning_rate": 0.0002490081252207941,
|
| 599 |
+
"loss": 1.6795,
|
| 600 |
+
"mean_token_accuracy": 0.6025200285613537,
|
| 601 |
"step": 37000
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 0.5095212538299014,
|
| 605 |
+
"grad_norm": 0.44578176736831665,
|
| 606 |
+
"learning_rate": 0.0002456112937851572,
|
| 607 |
+
"loss": 1.6703,
|
| 608 |
+
"mean_token_accuracy": 0.6039398565590381,
|
| 609 |
"step": 37500
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 0.5163148705476335,
|
| 613 |
+
"grad_norm": 0.45349687337875366,
|
| 614 |
+
"learning_rate": 0.00024221446234952039,
|
| 615 |
+
"loss": 1.6655,
|
| 616 |
+
"mean_token_accuracy": 0.6041326204538345,
|
| 617 |
"step": 38000
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 0.5231084872653654,
|
| 621 |
+
"grad_norm": 0.3655533194541931,
|
| 622 |
+
"learning_rate": 0.00023883121823962608,
|
| 623 |
+
"loss": 1.676,
|
| 624 |
+
"mean_token_accuracy": 0.6035989390015603,
|
| 625 |
"step": 38500
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 0.5299021039830974,
|
| 629 |
+
"grad_norm": 0.5141535401344299,
|
| 630 |
+
"learning_rate": 0.00023543438680398923,
|
| 631 |
+
"loss": 1.6784,
|
| 632 |
+
"mean_token_accuracy": 0.6029777365326882,
|
| 633 |
"step": 39000
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 0.5366957207008295,
|
| 637 |
+
"grad_norm": 0.3538263142108917,
|
| 638 |
+
"learning_rate": 0.00023205114269409495,
|
| 639 |
+
"loss": 1.666,
|
| 640 |
+
"mean_token_accuracy": 0.6050683797895908,
|
| 641 |
"step": 39500
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 0.5434893374185615,
|
| 645 |
+
"grad_norm": 0.4689236283302307,
|
| 646 |
+
"learning_rate": 0.00022865431125845812,
|
| 647 |
+
"loss": 1.6608,
|
| 648 |
+
"mean_token_accuracy": 0.6069498200118542,
|
| 649 |
"step": 40000
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.5502829541362936,
|
| 653 |
+
"grad_norm": 0.4244871735572815,
|
| 654 |
+
"learning_rate": 0.00022527106714856381,
|
| 655 |
+
"loss": 1.6701,
|
| 656 |
+
"mean_token_accuracy": 0.6025859090983867,
|
| 657 |
"step": 40500
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 0.5570765708540255,
|
| 661 |
+
"grad_norm": 0.49384674429893494,
|
| 662 |
+
"learning_rate": 0.00022187423571292698,
|
| 663 |
+
"loss": 1.6827,
|
| 664 |
+
"mean_token_accuracy": 0.6007110984623432,
|
| 665 |
"step": 41000
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 0.5638701875717576,
|
| 669 |
+
"grad_norm": 0.5254752039909363,
|
| 670 |
+
"learning_rate": 0.0002184909916030327,
|
| 671 |
+
"loss": 1.661,
|
| 672 |
+
"mean_token_accuracy": 0.6038983342051506,
|
| 673 |
"step": 41500
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 0.5706638042894896,
|
| 677 |
+
"grad_norm": 0.5031416416168213,
|
| 678 |
+
"learning_rate": 0.00021509416016739585,
|
| 679 |
+
"loss": 1.667,
|
| 680 |
+
"mean_token_accuracy": 0.6047138755917549,
|
| 681 |
"step": 42000
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 0.5774574210072216,
|
| 685 |
+
"grad_norm": 0.5616060495376587,
|
| 686 |
+
"learning_rate": 0.00021171091605750157,
|
| 687 |
+
"loss": 1.6731,
|
| 688 |
+
"mean_token_accuracy": 0.6030472720265388,
|
| 689 |
"step": 42500
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 0.5842510377249537,
|
| 693 |
+
"grad_norm": 0.6484994292259216,
|
| 694 |
+
"learning_rate": 0.00020831408462186474,
|
| 695 |
+
"loss": 1.6687,
|
| 696 |
+
"mean_token_accuracy": 0.6059929049313069,
|
| 697 |
"step": 43000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 0.5910446544426856,
|
| 701 |
+
"grad_norm": 0.6379068493843079,
|
| 702 |
+
"learning_rate": 0.00020493084051197044,
|
| 703 |
+
"loss": 1.6919,
|
| 704 |
+
"mean_token_accuracy": 0.5994226909875869,
|
| 705 |
"step": 43500
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.5978382711604177,
|
| 709 |
+
"grad_norm": 0.5151463150978088,
|
| 710 |
+
"learning_rate": 0.0002015340090763336,
|
| 711 |
+
"loss": 1.6743,
|
| 712 |
+
"mean_token_accuracy": 0.6019911893308163,
|
| 713 |
"step": 44000
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 0.6046318878781497,
|
| 717 |
+
"grad_norm": 0.4630017876625061,
|
| 718 |
+
"learning_rate": 0.0001981507649664393,
|
| 719 |
+
"loss": 1.6683,
|
| 720 |
+
"mean_token_accuracy": 0.6034873478114605,
|
| 721 |
"step": 44500
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 0.6114255045958817,
|
| 725 |
+
"grad_norm": 0.5501480102539062,
|
| 726 |
+
"learning_rate": 0.00019475393353080248,
|
| 727 |
+
"loss": 1.6675,
|
| 728 |
+
"mean_token_accuracy": 0.60292616507411,
|
| 729 |
"step": 45000
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 0.6182191213136137,
|
| 733 |
+
"grad_norm": 0.39942827820777893,
|
| 734 |
+
"learning_rate": 0.00019137068942090817,
|
| 735 |
+
"loss": 1.6877,
|
| 736 |
+
"mean_token_accuracy": 0.6008874903023242,
|
| 737 |
"step": 45500
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 0.6250127380313457,
|
| 741 |
+
"grad_norm": 0.4541276693344116,
|
| 742 |
+
"learning_rate": 0.00018797385798527134,
|
| 743 |
+
"loss": 1.6633,
|
| 744 |
+
"mean_token_accuracy": 0.6041645866334439,
|
| 745 |
"step": 46000
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 0.6318063547490778,
|
| 749 |
+
"grad_norm": 0.41248592734336853,
|
| 750 |
+
"learning_rate": 0.00018459061387537704,
|
| 751 |
+
"loss": 1.6802,
|
| 752 |
+
"mean_token_accuracy": 0.6013916251957416,
|
| 753 |
"step": 46500
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 0.6385999714668098,
|
| 757 |
+
"grad_norm": 0.5644799470901489,
|
| 758 |
+
"learning_rate": 0.0001811937824397402,
|
| 759 |
+
"loss": 1.6608,
|
| 760 |
+
"mean_token_accuracy": 0.6057685135900974,
|
| 761 |
"step": 47000
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.6453935881845418,
|
| 765 |
+
"grad_norm": 0.4525160789489746,
|
| 766 |
+
"learning_rate": 0.00017781053832984593,
|
| 767 |
+
"loss": 1.6779,
|
| 768 |
+
"mean_token_accuracy": 0.6037380565404892,
|
| 769 |
"step": 47500
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 0.6521872049022738,
|
| 773 |
+
"grad_norm": 0.516008198261261,
|
| 774 |
+
"learning_rate": 0.00017441370689420907,
|
| 775 |
+
"loss": 1.6733,
|
| 776 |
+
"mean_token_accuracy": 0.6011509810388088,
|
| 777 |
"step": 48000
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 0.6589808216200058,
|
| 781 |
+
"grad_norm": 0.5842276811599731,
|
| 782 |
+
"learning_rate": 0.0001710304627843148,
|
| 783 |
+
"loss": 1.6691,
|
| 784 |
+
"mean_token_accuracy": 0.6040294868052005,
|
| 785 |
"step": 48500
|
| 786 |
},
|
| 787 |
{
|
| 788 |
"epoch": 0.6657744383377379,
|
| 789 |
+
"grad_norm": 0.5395560264587402,
|
| 790 |
+
"learning_rate": 0.00016763363134867797,
|
| 791 |
+
"loss": 1.6584,
|
| 792 |
+
"mean_token_accuracy": 0.6059439519941807,
|
| 793 |
"step": 49000
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 0.6725680550554699,
|
| 797 |
+
"grad_norm": 0.4034149646759033,
|
| 798 |
+
"learning_rate": 0.00016425038723878366,
|
| 799 |
+
"loss": 1.6763,
|
| 800 |
+
"mean_token_accuracy": 0.6028572465479374,
|
| 801 |
"step": 49500
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 0.679361671773202,
|
| 805 |
+
"grad_norm": 0.4670972526073456,
|
| 806 |
+
"learning_rate": 0.00016085355580314683,
|
| 807 |
+
"loss": 1.6737,
|
| 808 |
+
"mean_token_accuracy": 0.6027044896781445,
|
| 809 |
"step": 50000
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 0.6861552884909339,
|
| 813 |
+
"grad_norm": 0.4919074475765228,
|
| 814 |
+
"learning_rate": 0.00015747031169325256,
|
| 815 |
+
"loss": 1.6724,
|
| 816 |
+
"mean_token_accuracy": 0.6025184873044491,
|
| 817 |
"step": 50500
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.6929489052086659,
|
| 821 |
+
"grad_norm": 0.5459266304969788,
|
| 822 |
+
"learning_rate": 0.0001540734802576157,
|
| 823 |
+
"loss": 1.6712,
|
| 824 |
+
"mean_token_accuracy": 0.601843435794115,
|
| 825 |
"step": 51000
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 0.699742521926398,
|
| 829 |
+
"grad_norm": 0.5790686011314392,
|
| 830 |
+
"learning_rate": 0.0001506902361477214,
|
| 831 |
+
"loss": 1.6766,
|
| 832 |
+
"mean_token_accuracy": 0.6018453196883201,
|
| 833 |
"step": 51500
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 0.70653613864413,
|
| 837 |
+
"grad_norm": 0.6184782981872559,
|
| 838 |
+
"learning_rate": 0.00014729340471208457,
|
| 839 |
+
"loss": 1.6605,
|
| 840 |
+
"mean_token_accuracy": 0.6039663677811623,
|
| 841 |
"step": 52000
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 0.7133297553618619,
|
| 845 |
+
"grad_norm": 0.46344733238220215,
|
| 846 |
+
"learning_rate": 0.0001439101606021903,
|
| 847 |
+
"loss": 1.6668,
|
| 848 |
+
"mean_token_accuracy": 0.6035030370950699,
|
| 849 |
"step": 52500
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.720123372079594,
|
| 853 |
+
"grad_norm": 0.4513992965221405,
|
| 854 |
+
"learning_rate": 0.00014051332916655343,
|
| 855 |
+
"loss": 1.681,
|
| 856 |
+
"mean_token_accuracy": 0.6013197962939739,
|
| 857 |
"step": 53000
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 0.726916988797326,
|
| 861 |
+
"grad_norm": 0.6055275797843933,
|
| 862 |
+
"learning_rate": 0.00013713008505665913,
|
| 863 |
+
"loss": 1.6645,
|
| 864 |
+
"mean_token_accuracy": 0.6029207228422165,
|
| 865 |
"step": 53500
|
| 866 |
},
|
| 867 |
{
|
| 868 |
"epoch": 0.7337106055150581,
|
| 869 |
+
"grad_norm": 0.7768318057060242,
|
| 870 |
+
"learning_rate": 0.00013373325362102232,
|
| 871 |
+
"loss": 1.6723,
|
| 872 |
+
"mean_token_accuracy": 0.6039300902187824,
|
| 873 |
"step": 54000
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.7405042222327901,
|
| 877 |
+
"grad_norm": 0.4635748863220215,
|
| 878 |
+
"learning_rate": 0.00013035000951112802,
|
| 879 |
+
"loss": 1.6755,
|
| 880 |
+
"mean_token_accuracy": 0.6006274480819702,
|
| 881 |
"step": 54500
|
| 882 |
},
|
| 883 |
{
|
| 884 |
"epoch": 0.747297838950522,
|
| 885 |
+
"grad_norm": 0.45187345147132874,
|
| 886 |
+
"learning_rate": 0.00012695317807549116,
|
| 887 |
+
"loss": 1.6575,
|
| 888 |
+
"mean_token_accuracy": 0.6050706767737866,
|
| 889 |
"step": 55000
|
| 890 |
},
|
| 891 |
{
|
| 892 |
"epoch": 0.7540914556682541,
|
| 893 |
+
"grad_norm": 0.4425111413002014,
|
| 894 |
+
"learning_rate": 0.0001235699339655969,
|
| 895 |
+
"loss": 1.6781,
|
| 896 |
+
"mean_token_accuracy": 0.6030424599051476,
|
| 897 |
"step": 55500
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 0.7608850723859861,
|
| 901 |
+
"grad_norm": 0.5505604147911072,
|
| 902 |
+
"learning_rate": 0.00012017310252996006,
|
| 903 |
+
"loss": 1.6554,
|
| 904 |
+
"mean_token_accuracy": 0.6084197340607643,
|
| 905 |
"step": 56000
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 0.7676786891037182,
|
| 909 |
+
"grad_norm": 0.5301482081413269,
|
| 910 |
+
"learning_rate": 0.00011678985842006575,
|
| 911 |
+
"loss": 1.6625,
|
| 912 |
+
"mean_token_accuracy": 0.602829116821289,
|
| 913 |
"step": 56500
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 0.7744723058214502,
|
| 917 |
+
"grad_norm": 0.5754754543304443,
|
| 918 |
+
"learning_rate": 0.00011339302698442892,
|
| 919 |
+
"loss": 1.6518,
|
| 920 |
+
"mean_token_accuracy": 0.6081080415248871,
|
| 921 |
"step": 57000
|
| 922 |
},
|
| 923 |
{
|
| 924 |
"epoch": 0.7812659225391821,
|
| 925 |
+
"grad_norm": 0.4835888743400574,
|
| 926 |
+
"learning_rate": 0.00011000978287453465,
|
| 927 |
+
"loss": 1.679,
|
| 928 |
+
"mean_token_accuracy": 0.6019827630519867,
|
| 929 |
"step": 57500
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 0.7880595392569142,
|
| 933 |
+
"grad_norm": 0.5146164894104004,
|
| 934 |
+
"learning_rate": 0.00010661295143889779,
|
| 935 |
+
"loss": 1.6735,
|
| 936 |
+
"mean_token_accuracy": 0.6020470056533813,
|
| 937 |
"step": 58000
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 0.7948531559746462,
|
| 941 |
+
"grad_norm": 0.5473788976669312,
|
| 942 |
+
"learning_rate": 0.00010322970732900351,
|
| 943 |
+
"loss": 1.6545,
|
| 944 |
+
"mean_token_accuracy": 0.6059571769535541,
|
| 945 |
"step": 58500
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 0.8016467726923783,
|
| 949 |
+
"grad_norm": 0.4318866729736328,
|
| 950 |
+
"learning_rate": 9.983287589336668e-05,
|
| 951 |
+
"loss": 1.6795,
|
| 952 |
+
"mean_token_accuracy": 0.6011920778155326,
|
| 953 |
"step": 59000
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 0.8084403894101102,
|
| 957 |
+
"grad_norm": 0.5313432216644287,
|
| 958 |
+
"learning_rate": 9.644963178347238e-05,
|
| 959 |
+
"loss": 1.6498,
|
| 960 |
+
"mean_token_accuracy": 0.6086946985721589,
|
| 961 |
"step": 59500
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 0.8152340061278422,
|
| 965 |
+
"grad_norm": 0.5233883261680603,
|
| 966 |
+
"learning_rate": 9.305280034783555e-05,
|
| 967 |
+
"loss": 1.663,
|
| 968 |
+
"mean_token_accuracy": 0.6036414765715599,
|
| 969 |
"step": 60000
|
| 970 |
},
|
| 971 |
{
|
| 972 |
"epoch": 0.8220276228455743,
|
| 973 |
+
"grad_norm": 0.5144417881965637,
|
| 974 |
+
"learning_rate": 8.966955623794124e-05,
|
| 975 |
+
"loss": 1.6813,
|
| 976 |
+
"mean_token_accuracy": 0.6005311135351658,
|
| 977 |
"step": 60500
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 0.8288212395633063,
|
| 981 |
+
"grad_norm": 0.563146710395813,
|
| 982 |
+
"learning_rate": 8.627272480230441e-05,
|
| 983 |
+
"loss": 1.6721,
|
| 984 |
+
"mean_token_accuracy": 0.6010738264620304,
|
| 985 |
"step": 61000
|
| 986 |
},
|
| 987 |
{
|
| 988 |
"epoch": 0.8356148562810384,
|
| 989 |
+
"grad_norm": 0.48985061049461365,
|
| 990 |
+
"learning_rate": 8.288948069241012e-05,
|
| 991 |
+
"loss": 1.6559,
|
| 992 |
+
"mean_token_accuracy": 0.6040105725228786,
|
| 993 |
"step": 61500
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 0.8424084729987703,
|
| 997 |
+
"grad_norm": 0.3702440559864044,
|
| 998 |
+
"learning_rate": 7.949264925677328e-05,
|
| 999 |
+
"loss": 1.6629,
|
| 1000 |
+
"mean_token_accuracy": 0.604276294618845,
|
| 1001 |
"step": 62000
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
"epoch": 0.8492020897165024,
|
| 1005 |
+
"grad_norm": 0.4425947368144989,
|
| 1006 |
+
"learning_rate": 7.610940514687899e-05,
|
| 1007 |
+
"loss": 1.6484,
|
| 1008 |
+
"mean_token_accuracy": 0.6068709415197372,
|
| 1009 |
"step": 62500
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
"epoch": 0.8559957064342344,
|
| 1013 |
+
"grad_norm": 0.4522407352924347,
|
| 1014 |
+
"learning_rate": 7.271257371124215e-05,
|
| 1015 |
+
"loss": 1.6567,
|
| 1016 |
+
"mean_token_accuracy": 0.6042509483098983,
|
| 1017 |
"step": 63000
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
"epoch": 0.8627893231519664,
|
| 1021 |
+
"grad_norm": 0.4882795512676239,
|
| 1022 |
+
"learning_rate": 6.932932960134787e-05,
|
| 1023 |
+
"loss": 1.6548,
|
| 1024 |
+
"mean_token_accuracy": 0.60615131264925,
|
| 1025 |
"step": 63500
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
"epoch": 0.8695829398696985,
|
| 1029 |
+
"grad_norm": 0.5515041351318359,
|
| 1030 |
+
"learning_rate": 6.593249816571103e-05,
|
| 1031 |
+
"loss": 1.663,
|
| 1032 |
+
"mean_token_accuracy": 0.6043142848014832,
|
| 1033 |
"step": 64000
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 0.8763765565874304,
|
| 1037 |
+
"grad_norm": 0.5676508545875549,
|
| 1038 |
+
"learning_rate": 6.254925405581674e-05,
|
| 1039 |
+
"loss": 1.6715,
|
| 1040 |
+
"mean_token_accuracy": 0.6038499280512333,
|
| 1041 |
"step": 64500
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 0.8831701733051625,
|
| 1045 |
+
"grad_norm": 0.6217424273490906,
|
| 1046 |
+
"learning_rate": 5.91524226201799e-05,
|
| 1047 |
+
"loss": 1.6616,
|
| 1048 |
+
"mean_token_accuracy": 0.6056457755565643,
|
| 1049 |
"step": 65000
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 0.8899637900228945,
|
| 1053 |
+
"grad_norm": 0.4277644157409668,
|
| 1054 |
+
"learning_rate": 5.576917851028561e-05,
|
| 1055 |
+
"loss": 1.6739,
|
| 1056 |
+
"mean_token_accuracy": 0.6015461843907833,
|
| 1057 |
"step": 65500
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 0.8967574067406265,
|
| 1061 |
+
"grad_norm": 0.5268669128417969,
|
| 1062 |
+
"learning_rate": 5.2372347074648765e-05,
|
| 1063 |
+
"loss": 1.6648,
|
| 1064 |
+
"mean_token_accuracy": 0.6022867387533188,
|
| 1065 |
"step": 66000
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
"epoch": 0.9035510234583585,
|
| 1069 |
+
"grad_norm": 0.4265219271183014,
|
| 1070 |
+
"learning_rate": 4.898910296475448e-05,
|
| 1071 |
+
"loss": 1.6598,
|
| 1072 |
+
"mean_token_accuracy": 0.6034302037358285,
|
| 1073 |
"step": 66500
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 0.9103446401760905,
|
| 1077 |
+
"grad_norm": 0.43360719084739685,
|
| 1078 |
+
"learning_rate": 4.5592271529117644e-05,
|
| 1079 |
+
"loss": 1.6673,
|
| 1080 |
+
"mean_token_accuracy": 0.6040239284336567,
|
| 1081 |
"step": 67000
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
"epoch": 0.9171382568938226,
|
| 1085 |
+
"grad_norm": 0.6076446771621704,
|
| 1086 |
+
"learning_rate": 4.220902741922335e-05,
|
| 1087 |
+
"loss": 1.6684,
|
| 1088 |
+
"mean_token_accuracy": 0.60401923969388,
|
| 1089 |
"step": 67500
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
"epoch": 0.9239318736115546,
|
| 1093 |
+
"grad_norm": 0.5381833910942078,
|
| 1094 |
+
"learning_rate": 3.881219598358651e-05,
|
| 1095 |
+
"loss": 1.6626,
|
| 1096 |
+
"mean_token_accuracy": 0.6033281030356884,
|
| 1097 |
"step": 68000
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
"epoch": 0.9307254903292866,
|
| 1101 |
+
"grad_norm": 0.5558338165283203,
|
| 1102 |
+
"learning_rate": 3.542895187369223e-05,
|
| 1103 |
+
"loss": 1.6664,
|
| 1104 |
+
"mean_token_accuracy": 0.6059176291823387,
|
| 1105 |
"step": 68500
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 0.9375191070470186,
|
| 1109 |
+
"grad_norm": 0.5438753962516785,
|
| 1110 |
+
"learning_rate": 3.2032120438055383e-05,
|
| 1111 |
+
"loss": 1.6533,
|
| 1112 |
+
"mean_token_accuracy": 0.6052996191084385,
|
| 1113 |
"step": 69000
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 0.9443127237647506,
|
| 1117 |
+
"grad_norm": 0.5837370753288269,
|
| 1118 |
+
"learning_rate": 2.8648876328161093e-05,
|
| 1119 |
+
"loss": 1.6775,
|
| 1120 |
+
"mean_token_accuracy": 0.6013322997689247,
|
| 1121 |
"step": 69500
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
"epoch": 0.9511063404824827,
|
| 1125 |
+
"grad_norm": 0.4310416877269745,
|
| 1126 |
+
"learning_rate": 2.5252044892524253e-05,
|
| 1127 |
+
"loss": 1.6548,
|
| 1128 |
+
"mean_token_accuracy": 0.6055889129638672,
|
| 1129 |
"step": 70000
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 0.9578999572002147,
|
| 1133 |
+
"grad_norm": 0.5146716237068176,
|
| 1134 |
+
"learning_rate": 2.1868800782629962e-05,
|
| 1135 |
+
"loss": 1.6663,
|
| 1136 |
+
"mean_token_accuracy": 0.6025079537630081,
|
| 1137 |
"step": 70500
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 0.9646935739179467,
|
| 1141 |
+
"grad_norm": 0.4616748094558716,
|
| 1142 |
+
"learning_rate": 1.8471969346993126e-05,
|
| 1143 |
+
"loss": 1.6703,
|
| 1144 |
+
"mean_token_accuracy": 0.603231570750475,
|
| 1145 |
"step": 71000
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 0.9714871906356787,
|
| 1149 |
+
"grad_norm": 0.4664286673069,
|
| 1150 |
+
"learning_rate": 1.5088725237098835e-05,
|
| 1151 |
+
"loss": 1.6628,
|
| 1152 |
+
"mean_token_accuracy": 0.6033680983781815,
|
| 1153 |
"step": 71500
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
"epoch": 0.9782808073534107,
|
| 1157 |
+
"grad_norm": 0.4937819242477417,
|
| 1158 |
+
"learning_rate": 1.1691893801461997e-05,
|
| 1159 |
+
"loss": 1.6668,
|
| 1160 |
+
"mean_token_accuracy": 0.6043202468752861,
|
| 1161 |
"step": 72000
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
"epoch": 0.9850744240711428,
|
| 1165 |
+
"grad_norm": 0.45690104365348816,
|
| 1166 |
+
"learning_rate": 8.308649691567706e-06,
|
| 1167 |
+
"loss": 1.66,
|
| 1168 |
+
"mean_token_accuracy": 0.6045957347154617,
|
| 1169 |
"step": 72500
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
"epoch": 0.9918680407888748,
|
| 1173 |
+
"grad_norm": 0.43818211555480957,
|
| 1174 |
+
"learning_rate": 4.911818255930868e-06,
|
| 1175 |
+
"loss": 1.6669,
|
| 1176 |
+
"mean_token_accuracy": 0.6046621772050857,
|
| 1177 |
"step": 73000
|
| 1178 |
},
|
| 1179 |
{
|
| 1180 |
"epoch": 0.9986616575066067,
|
| 1181 |
+
"grad_norm": 0.4940794110298157,
|
| 1182 |
+
"learning_rate": 1.528574146036577e-06,
|
| 1183 |
+
"loss": 1.6517,
|
| 1184 |
+
"mean_token_accuracy": 0.606719139367342,
|
| 1185 |
"step": 73500
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 0.9999932063832823,
|
| 1189 |
+
"mean_token_accuracy": 0.6021248713439825,
|
| 1190 |
"step": 73598,
|
| 1191 |
+
"total_flos": 9.078073041438638e+17,
|
| 1192 |
+
"train_loss": 1.6789688296530434,
|
| 1193 |
+
"train_runtime": 12675.9905,
|
| 1194 |
+
"train_samples_per_second": 23.224,
|
| 1195 |
"train_steps_per_second": 5.806
|
| 1196 |
}
|
| 1197 |
],
|
|
|
|
| 1212 |
"attributes": {}
|
| 1213 |
}
|
| 1214 |
},
|
| 1215 |
+
"total_flos": 9.078073041438638e+17,
|
| 1216 |
"train_batch_size": 1,
|
| 1217 |
"trial_name": null,
|
| 1218 |
"trial_params": null
|