Upload folder using huggingface_hub
Browse files- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +738 -102
- training_args.bin +1 -1
adapter_config.json
CHANGED
|
@@ -20,10 +20,10 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"qkv_proj",
|
| 24 |
-
"down_proj",
|
| 25 |
"o_proj",
|
| 26 |
-
"gate_up_proj"
|
|
|
|
|
|
|
| 27 |
],
|
| 28 |
"task_type": "CAUSAL_LM",
|
| 29 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
| 23 |
"o_proj",
|
| 24 |
+
"gate_up_proj",
|
| 25 |
+
"down_proj",
|
| 26 |
+
"qkv_proj"
|
| 27 |
],
|
| 28 |
"task_type": "CAUSAL_LM",
|
| 29 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 402688040
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17df5b37264eec43ee3b04c003030e9ef85cf5b6f14b7f60e70e79bb01519c0e
|
| 3 |
size 402688040
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 805522170
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:414621f021c20ee99d8749762fb235a920491413d965ecd572e7b6fd2c8676c4
|
| 3 |
size 805522170
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0954ce5829feaa075bd00d46c62f4c6b3adca283c9efdc99db72335353dd29ce
|
| 3 |
size 14244
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37332900e696ff0a98c0047cec7566becbfb436912055ab70393a526c6bb795d
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,235 +1,871 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "//outputs/task7_microsoft/Phi-3.5-mini-instruct/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.2962962962962963,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.5925925925925926,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.8888888888888888,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.0,
|
| 34 |
-
"eval_loss": 0.
|
| 35 |
-
"eval_runtime": 3.
|
| 36 |
"eval_samples_per_second": 4.429,
|
| 37 |
"eval_steps_per_second": 0.591,
|
| 38 |
"step": 34
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.1777777777777778,
|
| 42 |
-
"grad_norm": 0.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 40
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.474074074074074,
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 1.7703703703703704,
|
| 56 |
-
"grad_norm": 0.
|
| 57 |
-
"learning_rate": 9.
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 2.0,
|
| 63 |
-
"eval_loss": 0.
|
| 64 |
-
"eval_runtime": 3.
|
| 65 |
-
"eval_samples_per_second": 4.
|
| 66 |
-
"eval_steps_per_second": 0.
|
| 67 |
"step": 68
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 2.0592592592592593,
|
| 71 |
-
"grad_norm":
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 70
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 2.3555555555555556,
|
| 78 |
-
"grad_norm":
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 80
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.651851851851852,
|
| 85 |
-
"grad_norm": 0.
|
| 86 |
-
"learning_rate":
|
| 87 |
-
"loss": 0.
|
| 88 |
"step": 90
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 2.948148148148148,
|
| 92 |
-
"grad_norm": 0.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 3.0,
|
| 99 |
-
"eval_loss": 0.
|
| 100 |
-
"eval_runtime": 3.
|
| 101 |
-
"eval_samples_per_second": 4.
|
| 102 |
-
"eval_steps_per_second": 0.
|
| 103 |
"step": 102
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 3.237037037037037,
|
| 107 |
-
"grad_norm":
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 110
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 3.533333333333333,
|
| 114 |
-
"grad_norm": 0.
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 120
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 3.8296296296296295,
|
| 121 |
-
"grad_norm": 0.
|
| 122 |
-
"learning_rate":
|
| 123 |
-
"loss": 0.
|
| 124 |
"step": 130
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 4.0,
|
| 128 |
-
"eval_loss": 0.
|
| 129 |
-
"eval_runtime": 3.
|
| 130 |
-
"eval_samples_per_second": 4.
|
| 131 |
-
"eval_steps_per_second": 0.
|
| 132 |
"step": 136
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 4.118518518518519,
|
| 136 |
-
"grad_norm": 0.
|
| 137 |
-
"learning_rate":
|
| 138 |
-
"loss": 0.
|
| 139 |
"step": 140
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 4.4148148148148145,
|
| 143 |
-
"grad_norm": 0.
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 150
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 4.711111111111111,
|
| 150 |
-
"grad_norm": 0.
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 160
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 5.0,
|
| 157 |
-
"grad_norm": 0.
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss": 0.
|
| 160 |
"step": 170
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 5.0,
|
| 164 |
-
"eval_loss": 0.
|
| 165 |
-
"eval_runtime": 3.
|
| 166 |
-
"eval_samples_per_second": 4.
|
| 167 |
-
"eval_steps_per_second": 0.
|
| 168 |
"step": 170
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 5.296296296296296,
|
| 172 |
-
"grad_norm": 0.
|
| 173 |
-
"learning_rate":
|
| 174 |
-
"loss": 0.
|
| 175 |
"step": 180
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 5.592592592592593,
|
| 179 |
-
"grad_norm": 0.
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 190
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 5.888888888888889,
|
| 186 |
-
"grad_norm": 0.
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 200
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.0,
|
| 193 |
-
"eval_loss": 0.
|
| 194 |
-
"eval_runtime": 3.
|
| 195 |
-
"eval_samples_per_second": 4.
|
| 196 |
-
"eval_steps_per_second": 0.
|
| 197 |
"step": 204
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 6.177777777777778,
|
| 201 |
-
"grad_norm": 0.
|
| 202 |
-
"learning_rate":
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 210
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 6.474074074074074,
|
| 208 |
-
"grad_norm": 0.
|
| 209 |
-
"learning_rate":
|
| 210 |
-
"loss": 0.
|
| 211 |
"step": 220
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 6.770370370370371,
|
| 215 |
-
"grad_norm": 0.
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 230
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 7.0,
|
| 222 |
-
"eval_loss": 0.
|
| 223 |
-
"eval_runtime": 3.
|
| 224 |
-
"eval_samples_per_second": 4.
|
| 225 |
-
"eval_steps_per_second": 0.
|
| 226 |
"step": 238
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 227 |
}
|
| 228 |
],
|
| 229 |
"logging_steps": 10,
|
| 230 |
-
"max_steps":
|
| 231 |
"num_input_tokens_seen": 0,
|
| 232 |
-
"num_train_epochs":
|
| 233 |
"save_steps": 500,
|
| 234 |
"stateful_callbacks": {
|
| 235 |
"TrainerControl": {
|
|
@@ -243,7 +879,7 @@
|
|
| 243 |
"attributes": {}
|
| 244 |
}
|
| 245 |
},
|
| 246 |
-
"total_flos":
|
| 247 |
"train_batch_size": 1,
|
| 248 |
"trial_name": null,
|
| 249 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6473982334136963,
|
| 3 |
+
"best_model_checkpoint": "//outputs/task7_microsoft/Phi-3.5-mini-instruct/checkpoint-918",
|
| 4 |
+
"epoch": 27.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 918,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.2962962962962963,
|
| 13 |
+
"grad_norm": 0.9734601378440857,
|
| 14 |
+
"learning_rate": 2e-07,
|
| 15 |
+
"loss": 0.9542,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.5925925925925926,
|
| 20 |
+
"grad_norm": 1.0604201555252075,
|
| 21 |
+
"learning_rate": 4e-07,
|
| 22 |
+
"loss": 0.754,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.8888888888888888,
|
| 27 |
+
"grad_norm": 1.4953643083572388,
|
| 28 |
+
"learning_rate": 6e-07,
|
| 29 |
+
"loss": 0.8703,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.0,
|
| 34 |
+
"eval_loss": 0.7866833209991455,
|
| 35 |
+
"eval_runtime": 3.3869,
|
| 36 |
"eval_samples_per_second": 4.429,
|
| 37 |
"eval_steps_per_second": 0.591,
|
| 38 |
"step": 34
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.1777777777777778,
|
| 42 |
+
"grad_norm": 0.9286134243011475,
|
| 43 |
+
"learning_rate": 8e-07,
|
| 44 |
+
"loss": 0.7908,
|
| 45 |
"step": 40
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.474074074074074,
|
| 49 |
+
"grad_norm": 1.2561135292053223,
|
| 50 |
+
"learning_rate": 1e-06,
|
| 51 |
+
"loss": 0.9372,
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 1.7703703703703704,
|
| 56 |
+
"grad_norm": 0.749400794506073,
|
| 57 |
+
"learning_rate": 9.99726628670463e-07,
|
| 58 |
+
"loss": 0.7326,
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 2.0,
|
| 63 |
+
"eval_loss": 0.7615469098091125,
|
| 64 |
+
"eval_runtime": 3.3795,
|
| 65 |
+
"eval_samples_per_second": 4.439,
|
| 66 |
+
"eval_steps_per_second": 0.592,
|
| 67 |
"step": 68
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 2.0592592592592593,
|
| 71 |
+
"grad_norm": 1.545391321182251,
|
| 72 |
+
"learning_rate": 9.989068136093872e-07,
|
| 73 |
+
"loss": 0.8027,
|
| 74 |
"step": 70
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 2.3555555555555556,
|
| 78 |
+
"grad_norm": 1.2902804613113403,
|
| 79 |
+
"learning_rate": 9.975414512725056e-07,
|
| 80 |
+
"loss": 0.8375,
|
| 81 |
"step": 80
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.651851851851852,
|
| 85 |
+
"grad_norm": 0.8745124936103821,
|
| 86 |
+
"learning_rate": 9.956320346634875e-07,
|
| 87 |
+
"loss": 0.692,
|
| 88 |
"step": 90
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 2.948148148148148,
|
| 92 |
+
"grad_norm": 0.71682208776474,
|
| 93 |
+
"learning_rate": 9.931806517013612e-07,
|
| 94 |
+
"loss": 0.7837,
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 3.0,
|
| 99 |
+
"eval_loss": 0.7384690642356873,
|
| 100 |
+
"eval_runtime": 3.38,
|
| 101 |
+
"eval_samples_per_second": 4.438,
|
| 102 |
+
"eval_steps_per_second": 0.592,
|
| 103 |
"step": 102
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 3.237037037037037,
|
| 107 |
+
"grad_norm": 1.1133283376693726,
|
| 108 |
+
"learning_rate": 9.901899829374047e-07,
|
| 109 |
+
"loss": 0.8015,
|
| 110 |
"step": 110
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 3.533333333333333,
|
| 114 |
+
"grad_norm": 0.6814171075820923,
|
| 115 |
+
"learning_rate": 9.866632986240029e-07,
|
| 116 |
+
"loss": 0.7579,
|
| 117 |
"step": 120
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 3.8296296296296295,
|
| 121 |
+
"grad_norm": 0.6647155284881592,
|
| 122 |
+
"learning_rate": 9.826044551386742e-07,
|
| 123 |
+
"loss": 0.7589,
|
| 124 |
"step": 130
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 4.0,
|
| 128 |
+
"eval_loss": 0.7234861850738525,
|
| 129 |
+
"eval_runtime": 3.3808,
|
| 130 |
+
"eval_samples_per_second": 4.437,
|
| 131 |
+
"eval_steps_per_second": 0.592,
|
| 132 |
"step": 136
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 4.118518518518519,
|
| 136 |
+
"grad_norm": 0.41486606001853943,
|
| 137 |
+
"learning_rate": 9.780178907671788e-07,
|
| 138 |
+
"loss": 0.7408,
|
| 139 |
"step": 140
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 4.4148148148148145,
|
| 143 |
+
"grad_norm": 0.6310161352157593,
|
| 144 |
+
"learning_rate": 9.729086208503173e-07,
|
| 145 |
+
"loss": 0.7029,
|
| 146 |
"step": 150
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 4.711111111111111,
|
| 150 |
+
"grad_norm": 0.6634405851364136,
|
| 151 |
+
"learning_rate": 9.672822322997304e-07,
|
| 152 |
+
"loss": 0.7053,
|
| 153 |
"step": 160
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 5.0,
|
| 157 |
+
"grad_norm": 0.5732757449150085,
|
| 158 |
+
"learning_rate": 9.611448774886923e-07,
|
| 159 |
+
"loss": 0.8443,
|
| 160 |
"step": 170
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 5.0,
|
| 164 |
+
"eval_loss": 0.7118446230888367,
|
| 165 |
+
"eval_runtime": 3.3807,
|
| 166 |
+
"eval_samples_per_second": 4.437,
|
| 167 |
+
"eval_steps_per_second": 0.592,
|
| 168 |
"step": 170
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 5.296296296296296,
|
| 172 |
+
"grad_norm": 0.5060367584228516,
|
| 173 |
+
"learning_rate": 9.545032675245813e-07,
|
| 174 |
+
"loss": 0.7704,
|
| 175 |
"step": 180
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 5.592592592592593,
|
| 179 |
+
"grad_norm": 0.47104501724243164,
|
| 180 |
+
"learning_rate": 9.473646649103817e-07,
|
| 181 |
+
"loss": 0.758,
|
| 182 |
"step": 190
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 5.888888888888889,
|
| 186 |
+
"grad_norm": 0.6369937062263489,
|
| 187 |
+
"learning_rate": 9.397368756032444e-07,
|
| 188 |
+
"loss": 0.6357,
|
| 189 |
"step": 200
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.0,
|
| 193 |
+
"eval_loss": 0.7020046710968018,
|
| 194 |
+
"eval_runtime": 3.3979,
|
| 195 |
+
"eval_samples_per_second": 4.414,
|
| 196 |
+
"eval_steps_per_second": 0.589,
|
| 197 |
"step": 204
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 6.177777777777778,
|
| 201 |
+
"grad_norm": 0.5188702940940857,
|
| 202 |
+
"learning_rate": 9.316282404787869e-07,
|
| 203 |
+
"loss": 0.7171,
|
| 204 |
"step": 210
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 6.474074074074074,
|
| 208 |
+
"grad_norm": 0.5187780857086182,
|
| 209 |
+
"learning_rate": 9.230476262104676e-07,
|
| 210 |
+
"loss": 0.7533,
|
| 211 |
"step": 220
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 6.770370370370371,
|
| 215 |
+
"grad_norm": 0.3468983471393585,
|
| 216 |
+
"learning_rate": 9.1400441557401e-07,
|
| 217 |
+
"loss": 0.7398,
|
| 218 |
"step": 230
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 7.0,
|
| 222 |
+
"eval_loss": 0.694381058216095,
|
| 223 |
+
"eval_runtime": 3.3812,
|
| 224 |
+
"eval_samples_per_second": 4.436,
|
| 225 |
+
"eval_steps_per_second": 0.591,
|
| 226 |
"step": 238
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 7.059259259259259,
|
| 230 |
+
"grad_norm": 0.48491501808166504,
|
| 231 |
+
"learning_rate": 9.045084971874737e-07,
|
| 232 |
+
"loss": 0.6413,
|
| 233 |
+
"step": 240
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.355555555555555,
|
| 237 |
+
"grad_norm": 0.5511935949325562,
|
| 238 |
+
"learning_rate": 8.945702546981968e-07,
|
| 239 |
+
"loss": 0.7137,
|
| 240 |
+
"step": 250
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.651851851851852,
|
| 244 |
+
"grad_norm": 0.4468119144439697,
|
| 245 |
+
"learning_rate": 8.842005554284295e-07,
|
| 246 |
+
"loss": 0.7029,
|
| 247 |
+
"step": 260
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 7.948148148148148,
|
| 251 |
+
"grad_norm": 1.057121992111206,
|
| 252 |
+
"learning_rate": 8.734107384920769e-07,
|
| 253 |
+
"loss": 0.7468,
|
| 254 |
+
"step": 270
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 8.0,
|
| 258 |
+
"eval_loss": 0.6881054043769836,
|
| 259 |
+
"eval_runtime": 3.3795,
|
| 260 |
+
"eval_samples_per_second": 4.439,
|
| 261 |
+
"eval_steps_per_second": 0.592,
|
| 262 |
+
"step": 272
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 8.237037037037037,
|
| 266 |
+
"grad_norm": 0.6974695324897766,
|
| 267 |
+
"learning_rate": 8.622126023955445e-07,
|
| 268 |
+
"loss": 0.6952,
|
| 269 |
+
"step": 280
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 8.533333333333333,
|
| 273 |
+
"grad_norm": 0.5931522250175476,
|
| 274 |
+
"learning_rate": 8.506183921362442e-07,
|
| 275 |
+
"loss": 0.6434,
|
| 276 |
+
"step": 290
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 8.829629629629629,
|
| 280 |
+
"grad_norm": 0.5849316716194153,
|
| 281 |
+
"learning_rate": 8.386407858128706e-07,
|
| 282 |
+
"loss": 0.711,
|
| 283 |
+
"step": 300
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 9.0,
|
| 287 |
+
"eval_loss": 0.6817243695259094,
|
| 288 |
+
"eval_runtime": 3.3796,
|
| 289 |
+
"eval_samples_per_second": 4.438,
|
| 290 |
+
"eval_steps_per_second": 0.592,
|
| 291 |
+
"step": 306
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 9.118518518518519,
|
| 295 |
+
"grad_norm": 0.4516354501247406,
|
| 296 |
+
"learning_rate": 8.262928807620843e-07,
|
| 297 |
+
"loss": 0.7294,
|
| 298 |
+
"step": 310
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 9.414814814814815,
|
| 302 |
+
"grad_norm": 0.3970250189304352,
|
| 303 |
+
"learning_rate": 8.135881792367685e-07,
|
| 304 |
+
"loss": 0.6861,
|
| 305 |
+
"step": 320
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 9.71111111111111,
|
| 309 |
+
"grad_norm": 0.4954105317592621,
|
| 310 |
+
"learning_rate": 8.005405736415125e-07,
|
| 311 |
+
"loss": 0.6443,
|
| 312 |
+
"step": 330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 10.0,
|
| 316 |
+
"grad_norm": 0.9217901825904846,
|
| 317 |
+
"learning_rate": 7.871643313414718e-07,
|
| 318 |
+
"loss": 0.693,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 10.0,
|
| 323 |
+
"eval_loss": 0.6770989298820496,
|
| 324 |
+
"eval_runtime": 3.3796,
|
| 325 |
+
"eval_samples_per_second": 4.438,
|
| 326 |
+
"eval_steps_per_second": 0.592,
|
| 327 |
+
"step": 340
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 10.296296296296296,
|
| 331 |
+
"grad_norm": 0.6658220887184143,
|
| 332 |
+
"learning_rate": 7.734740790612136e-07,
|
| 333 |
+
"loss": 0.6673,
|
| 334 |
+
"step": 350
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 10.592592592592592,
|
| 338 |
+
"grad_norm": 1.5193367004394531,
|
| 339 |
+
"learning_rate": 7.594847868906076e-07,
|
| 340 |
+
"loss": 0.8039,
|
| 341 |
+
"step": 360
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 10.88888888888889,
|
| 345 |
+
"grad_norm": 0.3789392411708832,
|
| 346 |
+
"learning_rate": 7.452117519152541e-07,
|
| 347 |
+
"loss": 0.6556,
|
| 348 |
+
"step": 370
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 11.0,
|
| 352 |
+
"eval_loss": 0.6725327968597412,
|
| 353 |
+
"eval_runtime": 3.3798,
|
| 354 |
+
"eval_samples_per_second": 4.438,
|
| 355 |
+
"eval_steps_per_second": 0.592,
|
| 356 |
+
"step": 374
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 11.177777777777777,
|
| 360 |
+
"grad_norm": 0.5015827417373657,
|
| 361 |
+
"learning_rate": 7.306705814893439e-07,
|
| 362 |
+
"loss": 0.6501,
|
| 363 |
+
"step": 380
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 11.474074074074075,
|
| 367 |
+
"grad_norm": 0.3010861575603485,
|
| 368 |
+
"learning_rate": 7.158771761692464e-07,
|
| 369 |
+
"loss": 0.6209,
|
| 370 |
+
"step": 390
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 11.77037037037037,
|
| 374 |
+
"grad_norm": 0.8139746785163879,
|
| 375 |
+
"learning_rate": 7.008477123264847e-07,
|
| 376 |
+
"loss": 0.7087,
|
| 377 |
+
"step": 400
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.0,
|
| 381 |
+
"eval_loss": 0.6691569685935974,
|
| 382 |
+
"eval_runtime": 3.3801,
|
| 383 |
+
"eval_samples_per_second": 4.438,
|
| 384 |
+
"eval_steps_per_second": 0.592,
|
| 385 |
+
"step": 408
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 12.059259259259258,
|
| 389 |
+
"grad_norm": 0.5303260087966919,
|
| 390 |
+
"learning_rate": 6.855986244591103e-07,
|
| 391 |
+
"loss": 0.6422,
|
| 392 |
+
"step": 410
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 12.355555555555556,
|
| 396 |
+
"grad_norm": 0.6748088002204895,
|
| 397 |
+
"learning_rate": 6.701465872208216e-07,
|
| 398 |
+
"loss": 0.6324,
|
| 399 |
+
"step": 420
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 12.651851851851852,
|
| 403 |
+
"grad_norm": 0.30257901549339294,
|
| 404 |
+
"learning_rate": 6.545084971874736e-07,
|
| 405 |
+
"loss": 0.6435,
|
| 406 |
+
"step": 430
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 12.948148148148148,
|
| 410 |
+
"grad_norm": 0.502646803855896,
|
| 411 |
+
"learning_rate": 6.387014543809223e-07,
|
| 412 |
+
"loss": 0.7937,
|
| 413 |
+
"step": 440
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 13.0,
|
| 417 |
+
"eval_loss": 0.6658554077148438,
|
| 418 |
+
"eval_runtime": 3.3771,
|
| 419 |
+
"eval_samples_per_second": 4.442,
|
| 420 |
+
"eval_steps_per_second": 0.592,
|
| 421 |
+
"step": 442
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 13.237037037037037,
|
| 425 |
+
"grad_norm": 0.5977727174758911,
|
| 426 |
+
"learning_rate": 6.227427435703995e-07,
|
| 427 |
+
"loss": 0.7328,
|
| 428 |
+
"step": 450
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 13.533333333333333,
|
| 432 |
+
"grad_norm": 0.5516536235809326,
|
| 433 |
+
"learning_rate": 6.066498153718734e-07,
|
| 434 |
+
"loss": 0.6106,
|
| 435 |
+
"step": 460
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 13.829629629629629,
|
| 439 |
+
"grad_norm": 0.6513301730155945,
|
| 440 |
+
"learning_rate": 5.90440267166055e-07,
|
| 441 |
+
"loss": 0.6848,
|
| 442 |
+
"step": 470
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 14.0,
|
| 446 |
+
"eval_loss": 0.6627060174942017,
|
| 447 |
+
"eval_runtime": 3.3775,
|
| 448 |
+
"eval_samples_per_second": 4.441,
|
| 449 |
+
"eval_steps_per_second": 0.592,
|
| 450 |
+
"step": 476
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 14.118518518518519,
|
| 454 |
+
"grad_norm": 0.548687756061554,
|
| 455 |
+
"learning_rate": 5.741318238559209e-07,
|
| 456 |
+
"loss": 0.6687,
|
| 457 |
+
"step": 480
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 14.414814814814815,
|
| 461 |
+
"grad_norm": 0.5744287967681885,
|
| 462 |
+
"learning_rate": 5.577423184847931e-07,
|
| 463 |
+
"loss": 0.6226,
|
| 464 |
+
"step": 490
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 14.71111111111111,
|
| 468 |
+
"grad_norm": 0.6195695996284485,
|
| 469 |
+
"learning_rate": 5.412896727361662e-07,
|
| 470 |
+
"loss": 0.6681,
|
| 471 |
+
"step": 500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 15.0,
|
| 475 |
+
"grad_norm": 0.9363336563110352,
|
| 476 |
+
"learning_rate": 5.247918773366111e-07,
|
| 477 |
+
"loss": 0.6825,
|
| 478 |
+
"step": 510
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 15.0,
|
| 482 |
+
"eval_loss": 0.6597088575363159,
|
| 483 |
+
"eval_runtime": 3.3779,
|
| 484 |
+
"eval_samples_per_second": 4.441,
|
| 485 |
+
"eval_steps_per_second": 0.592,
|
| 486 |
+
"step": 510
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 15.296296296296296,
|
| 490 |
+
"grad_norm": 0.4059470295906067,
|
| 491 |
+
"learning_rate": 5.082669723831793e-07,
|
| 492 |
+
"loss": 0.5709,
|
| 493 |
+
"step": 520
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 15.592592592592592,
|
| 497 |
+
"grad_norm": 0.5535906553268433,
|
| 498 |
+
"learning_rate": 4.917330276168208e-07,
|
| 499 |
+
"loss": 0.712,
|
| 500 |
+
"step": 530
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 15.88888888888889,
|
| 504 |
+
"grad_norm": 0.5456708669662476,
|
| 505 |
+
"learning_rate": 4.752081226633888e-07,
|
| 506 |
+
"loss": 0.6777,
|
| 507 |
+
"step": 540
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 16.0,
|
| 511 |
+
"eval_loss": 0.6580029726028442,
|
| 512 |
+
"eval_runtime": 3.3796,
|
| 513 |
+
"eval_samples_per_second": 4.438,
|
| 514 |
+
"eval_steps_per_second": 0.592,
|
| 515 |
+
"step": 544
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 16.177777777777777,
|
| 519 |
+
"grad_norm": 0.2641673684120178,
|
| 520 |
+
"learning_rate": 4.5871032726383385e-07,
|
| 521 |
+
"loss": 0.7006,
|
| 522 |
+
"step": 550
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 16.474074074074075,
|
| 526 |
+
"grad_norm": 0.49657076597213745,
|
| 527 |
+
"learning_rate": 4.4225768151520694e-07,
|
| 528 |
+
"loss": 0.7039,
|
| 529 |
+
"step": 560
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 16.77037037037037,
|
| 533 |
+
"grad_norm": 0.5019007325172424,
|
| 534 |
+
"learning_rate": 4.258681761440789e-07,
|
| 535 |
+
"loss": 0.7018,
|
| 536 |
+
"step": 570
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 17.0,
|
| 540 |
+
"eval_loss": 0.655741810798645,
|
| 541 |
+
"eval_runtime": 3.3817,
|
| 542 |
+
"eval_samples_per_second": 4.436,
|
| 543 |
+
"eval_steps_per_second": 0.591,
|
| 544 |
+
"step": 578
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 17.05925925925926,
|
| 548 |
+
"grad_norm": 0.5443792343139648,
|
| 549 |
+
"learning_rate": 4.095597328339452e-07,
|
| 550 |
+
"loss": 0.5107,
|
| 551 |
+
"step": 580
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 17.355555555555554,
|
| 555 |
+
"grad_norm": 0.6433914303779602,
|
| 556 |
+
"learning_rate": 3.9335018462812664e-07,
|
| 557 |
+
"loss": 0.724,
|
| 558 |
+
"step": 590
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 17.651851851851852,
|
| 562 |
+
"grad_norm": 0.8970219492912292,
|
| 563 |
+
"learning_rate": 3.772572564296004e-07,
|
| 564 |
+
"loss": 0.6995,
|
| 565 |
+
"step": 600
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 17.94814814814815,
|
| 569 |
+
"grad_norm": 0.43578681349754333,
|
| 570 |
+
"learning_rate": 3.612985456190778e-07,
|
| 571 |
+
"loss": 0.6126,
|
| 572 |
+
"step": 610
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 18.0,
|
| 576 |
+
"eval_loss": 0.6544412970542908,
|
| 577 |
+
"eval_runtime": 3.3793,
|
| 578 |
+
"eval_samples_per_second": 4.439,
|
| 579 |
+
"eval_steps_per_second": 0.592,
|
| 580 |
+
"step": 612
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 18.237037037037037,
|
| 584 |
+
"grad_norm": 0.6419870257377625,
|
| 585 |
+
"learning_rate": 3.454915028125263e-07,
|
| 586 |
+
"loss": 0.5766,
|
| 587 |
+
"step": 620
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 18.533333333333335,
|
| 591 |
+
"grad_norm": 0.3237641751766205,
|
| 592 |
+
"learning_rate": 3.2985341277917846e-07,
|
| 593 |
+
"loss": 0.6299,
|
| 594 |
+
"step": 630
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 18.82962962962963,
|
| 598 |
+
"grad_norm": 0.409332275390625,
|
| 599 |
+
"learning_rate": 3.1440137554088953e-07,
|
| 600 |
+
"loss": 0.5755,
|
| 601 |
+
"step": 640
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 19.0,
|
| 605 |
+
"eval_loss": 0.6529051661491394,
|
| 606 |
+
"eval_runtime": 3.3802,
|
| 607 |
+
"eval_samples_per_second": 4.438,
|
| 608 |
+
"eval_steps_per_second": 0.592,
|
| 609 |
+
"step": 646
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 19.118518518518517,
|
| 613 |
+
"grad_norm": 0.4559795558452606,
|
| 614 |
+
"learning_rate": 2.9915228767351535e-07,
|
| 615 |
+
"loss": 0.7702,
|
| 616 |
+
"step": 650
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 19.414814814814815,
|
| 620 |
+
"grad_norm": 0.44174766540527344,
|
| 621 |
+
"learning_rate": 2.841228238307536e-07,
|
| 622 |
+
"loss": 0.5212,
|
| 623 |
+
"step": 660
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 19.711111111111112,
|
| 627 |
+
"grad_norm": 0.5898323655128479,
|
| 628 |
+
"learning_rate": 2.6932941851065615e-07,
|
| 629 |
+
"loss": 0.6807,
|
| 630 |
+
"step": 670
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 20.0,
|
| 634 |
+
"grad_norm": 0.6098238825798035,
|
| 635 |
+
"learning_rate": 2.547882480847461e-07,
|
| 636 |
+
"loss": 0.6458,
|
| 637 |
+
"step": 680
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 20.0,
|
| 641 |
+
"eval_loss": 0.6510729193687439,
|
| 642 |
+
"eval_runtime": 3.3804,
|
| 643 |
+
"eval_samples_per_second": 4.437,
|
| 644 |
+
"eval_steps_per_second": 0.592,
|
| 645 |
+
"step": 680
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 20.296296296296298,
|
| 649 |
+
"grad_norm": 0.5178992748260498,
|
| 650 |
+
"learning_rate": 2.4051521310939254e-07,
|
| 651 |
+
"loss": 0.6025,
|
| 652 |
+
"step": 690
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 20.59259259259259,
|
| 656 |
+
"grad_norm": 0.7392374873161316,
|
| 657 |
+
"learning_rate": 2.2652592093878665e-07,
|
| 658 |
+
"loss": 0.7731,
|
| 659 |
+
"step": 700
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 20.88888888888889,
|
| 663 |
+
"grad_norm": 0.34529829025268555,
|
| 664 |
+
"learning_rate": 2.128356686585282e-07,
|
| 665 |
+
"loss": 0.6715,
|
| 666 |
+
"step": 710
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 21.0,
|
| 670 |
+
"eval_loss": 0.6509791016578674,
|
| 671 |
+
"eval_runtime": 3.38,
|
| 672 |
+
"eval_samples_per_second": 4.438,
|
| 673 |
+
"eval_steps_per_second": 0.592,
|
| 674 |
+
"step": 714
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 21.177777777777777,
|
| 678 |
+
"grad_norm": 0.5632665753364563,
|
| 679 |
+
"learning_rate": 1.9945942635848745e-07,
|
| 680 |
+
"loss": 0.65,
|
| 681 |
+
"step": 720
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 21.474074074074075,
|
| 685 |
+
"grad_norm": 0.41342058777809143,
|
| 686 |
+
"learning_rate": 1.8641182076323148e-07,
|
| 687 |
+
"loss": 0.6272,
|
| 688 |
+
"step": 730
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 21.77037037037037,
|
| 692 |
+
"grad_norm": 0.3611527383327484,
|
| 693 |
+
"learning_rate": 1.7370711923791564e-07,
|
| 694 |
+
"loss": 0.6173,
|
| 695 |
+
"step": 740
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 22.0,
|
| 699 |
+
"eval_loss": 0.649942934513092,
|
| 700 |
+
"eval_runtime": 3.38,
|
| 701 |
+
"eval_samples_per_second": 4.438,
|
| 702 |
+
"eval_steps_per_second": 0.592,
|
| 703 |
+
"step": 748
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 22.05925925925926,
|
| 707 |
+
"grad_norm": 0.9855077266693115,
|
| 708 |
+
"learning_rate": 1.6135921418712955e-07,
|
| 709 |
+
"loss": 0.6918,
|
| 710 |
+
"step": 750
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 22.355555555555554,
|
| 714 |
+
"grad_norm": 0.35098254680633545,
|
| 715 |
+
"learning_rate": 1.493816078637557e-07,
|
| 716 |
+
"loss": 0.5533,
|
| 717 |
+
"step": 760
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 22.651851851851852,
|
| 721 |
+
"grad_norm": 0.5630102157592773,
|
| 722 |
+
"learning_rate": 1.3778739760445552e-07,
|
| 723 |
+
"loss": 0.5668,
|
| 724 |
+
"step": 770
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 22.94814814814815,
|
| 728 |
+
"grad_norm": 0.42931997776031494,
|
| 729 |
+
"learning_rate": 1.2658926150792322e-07,
|
| 730 |
+
"loss": 0.7773,
|
| 731 |
+
"step": 780
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 23.0,
|
| 735 |
+
"eval_loss": 0.6492455005645752,
|
| 736 |
+
"eval_runtime": 3.3795,
|
| 737 |
+
"eval_samples_per_second": 4.439,
|
| 738 |
+
"eval_steps_per_second": 0.592,
|
| 739 |
+
"step": 782
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 23.237037037037037,
|
| 743 |
+
"grad_norm": 1.0845577716827393,
|
| 744 |
+
"learning_rate": 1.1579944457157059e-07,
|
| 745 |
+
"loss": 0.5997,
|
| 746 |
+
"step": 790
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 23.533333333333335,
|
| 750 |
+
"grad_norm": 0.8447713255882263,
|
| 751 |
+
"learning_rate": 1.0542974530180327e-07,
|
| 752 |
+
"loss": 0.6652,
|
| 753 |
+
"step": 800
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 23.82962962962963,
|
| 757 |
+
"grad_norm": 0.7863607406616211,
|
| 758 |
+
"learning_rate": 9.549150281252632e-08,
|
| 759 |
+
"loss": 0.5996,
|
| 760 |
+
"step": 810
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 24.0,
|
| 764 |
+
"eval_loss": 0.6488202810287476,
|
| 765 |
+
"eval_runtime": 3.3801,
|
| 766 |
+
"eval_samples_per_second": 4.438,
|
| 767 |
+
"eval_steps_per_second": 0.592,
|
| 768 |
+
"step": 816
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 24.118518518518517,
|
| 772 |
+
"grad_norm": 0.9356284141540527,
|
| 773 |
+
"learning_rate": 8.599558442598998e-08,
|
| 774 |
+
"loss": 0.7435,
|
| 775 |
+
"step": 820
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 24.414814814814815,
|
| 779 |
+
"grad_norm": 0.4750384986400604,
|
| 780 |
+
"learning_rate": 7.695237378953224e-08,
|
| 781 |
+
"loss": 0.6673,
|
| 782 |
+
"step": 830
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 24.711111111111112,
|
| 786 |
+
"grad_norm": 0.5514878630638123,
|
| 787 |
+
"learning_rate": 6.837175952121304e-08,
|
| 788 |
+
"loss": 0.6226,
|
| 789 |
+
"step": 840
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 25.0,
|
| 793 |
+
"grad_norm": 0.5611656904220581,
|
| 794 |
+
"learning_rate": 6.026312439675551e-08,
|
| 795 |
+
"loss": 0.5958,
|
| 796 |
+
"step": 850
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 25.0,
|
| 800 |
+
"eval_loss": 0.6484441161155701,
|
| 801 |
+
"eval_runtime": 3.3794,
|
| 802 |
+
"eval_samples_per_second": 4.439,
|
| 803 |
+
"eval_steps_per_second": 0.592,
|
| 804 |
+
"step": 850
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 25.296296296296298,
|
| 808 |
+
"grad_norm": 0.3849276900291443,
|
| 809 |
+
"learning_rate": 5.263533508961826e-08,
|
| 810 |
+
"loss": 0.7222,
|
| 811 |
+
"step": 860
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 25.59259259259259,
|
| 815 |
+
"grad_norm": 0.5683622360229492,
|
| 816 |
+
"learning_rate": 4.549673247541874e-08,
|
| 817 |
+
"loss": 0.7069,
|
| 818 |
+
"step": 870
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 25.88888888888889,
|
| 822 |
+
"grad_norm": 0.7111372351646423,
|
| 823 |
+
"learning_rate": 3.8855122511307626e-08,
|
| 824 |
+
"loss": 0.5773,
|
| 825 |
+
"step": 880
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 26.0,
|
| 829 |
+
"eval_loss": 0.6479985117912292,
|
| 830 |
+
"eval_runtime": 3.3801,
|
| 831 |
+
"eval_samples_per_second": 4.438,
|
| 832 |
+
"eval_steps_per_second": 0.592,
|
| 833 |
+
"step": 884
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 26.177777777777777,
|
| 837 |
+
"grad_norm": 0.27255579829216003,
|
| 838 |
+
"learning_rate": 3.271776770026963e-08,
|
| 839 |
+
"loss": 0.6349,
|
| 840 |
+
"step": 890
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 26.474074074074075,
|
| 844 |
+
"grad_norm": 0.54938143491745,
|
| 845 |
+
"learning_rate": 2.7091379149682682e-08,
|
| 846 |
+
"loss": 0.6322,
|
| 847 |
+
"step": 900
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 26.77037037037037,
|
| 851 |
+
"grad_norm": 0.6063421964645386,
|
| 852 |
+
"learning_rate": 2.1982109232821176e-08,
|
| 853 |
+
"loss": 0.6462,
|
| 854 |
+
"step": 910
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 27.0,
|
| 858 |
+
"eval_loss": 0.6473982334136963,
|
| 859 |
+
"eval_runtime": 3.3791,
|
| 860 |
+
"eval_samples_per_second": 4.439,
|
| 861 |
+
"eval_steps_per_second": 0.592,
|
| 862 |
+
"step": 918
|
| 863 |
}
|
| 864 |
],
|
| 865 |
"logging_steps": 10,
|
| 866 |
+
"max_steps": 1000,
|
| 867 |
"num_input_tokens_seen": 0,
|
| 868 |
+
"num_train_epochs": 31,
|
| 869 |
"save_steps": 500,
|
| 870 |
"stateful_callbacks": {
|
| 871 |
"TrainerControl": {
|
|
|
|
| 879 |
"attributes": {}
|
| 880 |
}
|
| 881 |
},
|
| 882 |
+
"total_flos": 8.836917475838976e+16,
|
| 883 |
"train_batch_size": 1,
|
| 884 |
"trial_name": null,
|
| 885 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5624
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a70798789a008d4e58669a5b402559307877fa0be790a9160ed0e412b4cc179
|
| 3 |
size 5624
|