Upload folder using huggingface_hub
Browse files- README.md +5 -5
- adapter_config.json +7 -7
- adapter_model.bin +1 -1
- checkpoint-126/adapter_config.json +7 -7
- checkpoint-126/adapter_model.safetensors +1 -1
- checkpoint-126/global_step126/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +1 -1
- checkpoint-126/global_step126/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +1 -1
- checkpoint-126/global_step126/mp_rank_00_model_states.pt +1 -1
- checkpoint-126/rng_state_0.pth +1 -1
- checkpoint-126/rng_state_1.pth +1 -1
- checkpoint-126/trainer_state.json +258 -258
- checkpoint-126/training_args.bin +1 -1
- checkpoint-168/adapter_config.json +7 -7
- checkpoint-168/adapter_model.safetensors +1 -1
- checkpoint-168/global_step168/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +1 -1
- checkpoint-168/global_step168/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +1 -1
- checkpoint-168/global_step168/mp_rank_00_model_states.pt +1 -1
- checkpoint-168/rng_state_0.pth +1 -1
- checkpoint-168/rng_state_1.pth +1 -1
- checkpoint-168/trainer_state.json +344 -344
- checkpoint-168/training_args.bin +1 -1
- config.json +17 -2
- tokenizer_config.json +1 -0
README.md
CHANGED
|
@@ -95,7 +95,7 @@ special_tokens:
|
|
| 95 |
|
| 96 |
This model is a fine-tuned version of [ai21labs/Jamba-v0.1](https://huggingface.co/ai21labs/Jamba-v0.1) on the None dataset.
|
| 97 |
It achieves the following results on the evaluation set:
|
| 98 |
-
- Loss: 0.
|
| 99 |
|
| 100 |
## Model description
|
| 101 |
|
|
@@ -133,10 +133,10 @@ The following hyperparameters were used during training:
|
|
| 133 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 134 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 135 |
| 0.6274 | 0.01 | 1 | 1.0298 |
|
| 136 |
-
| 0.
|
| 137 |
-
| 0.
|
| 138 |
-
| 0.
|
| 139 |
-
| 0.
|
| 140 |
|
| 141 |
|
| 142 |
### Framework versions
|
|
|
|
| 95 |
|
| 96 |
This model is a fine-tuned version of [ai21labs/Jamba-v0.1](https://huggingface.co/ai21labs/Jamba-v0.1) on the None dataset.
|
| 97 |
It achieves the following results on the evaluation set:
|
| 98 |
+
- Loss: 0.9641
|
| 99 |
|
| 100 |
## Model description
|
| 101 |
|
|
|
|
| 133 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 134 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 135 |
| 0.6274 | 0.01 | 1 | 1.0298 |
|
| 136 |
+
| 0.44 | 0.25 | 42 | 0.9770 |
|
| 137 |
+
| 0.4406 | 0.5 | 84 | 0.9653 |
|
| 138 |
+
| 0.4445 | 0.75 | 126 | 0.9645 |
|
| 139 |
+
| 0.4609 | 1.0 | 168 | 0.9641 |
|
| 140 |
|
| 141 |
|
| 142 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -20,17 +20,17 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"
|
| 24 |
-
"o_proj",
|
| 25 |
-
"up_proj",
|
| 26 |
"v_proj",
|
| 27 |
-
"down_proj",
|
| 28 |
"dt_proj",
|
| 29 |
-
"
|
| 30 |
-
"q_proj",
|
| 31 |
"in_proj",
|
|
|
|
|
|
|
|
|
|
| 32 |
"gate_proj",
|
| 33 |
-
"
|
|
|
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
+
"q_proj",
|
|
|
|
|
|
|
| 24 |
"v_proj",
|
|
|
|
| 25 |
"dt_proj",
|
| 26 |
+
"out_proj",
|
|
|
|
| 27 |
"in_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"x_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
"gate_proj",
|
| 32 |
+
"down_proj",
|
| 33 |
+
"up_proj",
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 531653306
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5666b45b26616be53d976f599b2d2d6262c975237daca20b37002cb90b38e7c
|
| 3 |
size 531653306
|
checkpoint-126/adapter_config.json
CHANGED
|
@@ -20,17 +20,17 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"
|
| 24 |
-
"o_proj",
|
| 25 |
-
"up_proj",
|
| 26 |
"v_proj",
|
| 27 |
-
"down_proj",
|
| 28 |
"dt_proj",
|
| 29 |
-
"
|
| 30 |
-
"q_proj",
|
| 31 |
"in_proj",
|
|
|
|
|
|
|
|
|
|
| 32 |
"gate_proj",
|
| 33 |
-
"
|
|
|
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
+
"q_proj",
|
|
|
|
|
|
|
| 24 |
"v_proj",
|
|
|
|
| 25 |
"dt_proj",
|
| 26 |
+
"out_proj",
|
|
|
|
| 27 |
"in_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"x_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
"gate_proj",
|
| 32 |
+
"down_proj",
|
| 33 |
+
"up_proj",
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
checkpoint-126/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 531611600
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e5244d4a10dbd753d970890412e3ca9603e7dcf433636b5774a56cd24230ca1
|
| 3 |
size 531611600
|
checkpoint-126/global_step126/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 797643792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30e35823423f4d40789f18071347d83fb7bfc4ac817c60dd4a4a37af4433ed7a
|
| 3 |
size 797643792
|
checkpoint-126/global_step126/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 797644432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bb2c01e320ef6c12ebd1a5ff11884d226fce0fc7a7455216a45c07ed35197d8
|
| 3 |
size 797644432
|
checkpoint-126/global_step126/mp_rank_00_model_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1345660121
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa91f679f40d2c52f881c918ddc04e8bf7639c32c1ebb9c47161e35088f9ff18
|
| 3 |
size 1345660121
|
checkpoint-126/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:136c534e0ab55c33db36cd2e9ff890cf7a74c156913de6ef32dd097847df2202
|
| 3 |
size 14512
|
checkpoint-126/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7e5b7eb90da0157ab2ee2a702d0793eaa00e971e5110cd0820814aa3fa7d061
|
| 3 |
size 14512
|
checkpoint-126/trainer_state.json
CHANGED
|
@@ -10,7 +10,7 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
"loss": 0.6274,
|
| 16 |
"step": 1
|
|
@@ -18,908 +18,908 @@
|
|
| 18 |
{
|
| 19 |
"epoch": 0.01,
|
| 20 |
"eval_loss": 1.0297596454620361,
|
| 21 |
-
"eval_runtime":
|
| 22 |
-
"eval_samples_per_second": 1.
|
| 23 |
-
"eval_steps_per_second": 0.
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.01,
|
| 28 |
-
"grad_norm": 0.
|
| 29 |
"learning_rate": 4e-05,
|
| 30 |
"loss": 0.6362,
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.02,
|
| 35 |
-
"grad_norm": 0.
|
| 36 |
"learning_rate": 6e-05,
|
| 37 |
"loss": 0.6299,
|
| 38 |
"step": 3
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.02,
|
| 42 |
-
"grad_norm": 0.
|
| 43 |
"learning_rate": 8e-05,
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 4
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.03,
|
| 49 |
-
"grad_norm": 0.
|
| 50 |
"learning_rate": 0.0001,
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 5
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.04,
|
| 56 |
-
"grad_norm": 0.
|
| 57 |
"learning_rate": 0.00012,
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 6
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.04,
|
| 63 |
-
"grad_norm": 0.
|
| 64 |
"learning_rate": 0.00014,
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.05,
|
| 70 |
-
"grad_norm": 0.
|
| 71 |
"learning_rate": 0.00016,
|
| 72 |
-
"loss": 0.
|
| 73 |
"step": 8
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.05,
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
"learning_rate": 0.00018,
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 9
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.06,
|
| 84 |
-
"grad_norm": 0.
|
| 85 |
"learning_rate": 0.0002,
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 10
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.07,
|
| 91 |
-
"grad_norm": 0.
|
| 92 |
"learning_rate": 0.00019998023297700658,
|
| 93 |
-
"loss": 0.
|
| 94 |
"step": 11
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.07,
|
| 98 |
-
"grad_norm": 0.
|
| 99 |
"learning_rate": 0.00019992093972273018,
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 12
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.08,
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
"learning_rate": 0.00019982214367819328,
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 13
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.08,
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
"learning_rate": 0.0001996838839014696,
|
| 114 |
"loss": 0.5495,
|
| 115 |
"step": 14
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.09,
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
"learning_rate": 0.00019950621505224273,
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 15
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.1,
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
"learning_rate": 0.00019928920737019733,
|
| 128 |
"loss": 0.5083,
|
| 129 |
"step": 16
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.1,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
"learning_rate": 0.0001990329466472502,
|
| 135 |
-
"loss": 0.
|
| 136 |
"step": 17
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.11,
|
| 140 |
-
"grad_norm": 0.
|
| 141 |
"learning_rate": 0.00019873753419363336,
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 18
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.11,
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
"learning_rate": 0.00019840308679784207,
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 19
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.12,
|
| 154 |
-
"grad_norm": 0.
|
| 155 |
"learning_rate": 0.00019802973668046363,
|
| 156 |
-
"loss": 0.
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.12,
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
"learning_rate": 0.0001976176314419051,
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 21
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.13,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
"learning_rate": 0.000197166934004041,
|
| 170 |
"loss": 0.4819,
|
| 171 |
"step": 22
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.14,
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
"learning_rate": 0.00019667782254580374,
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 23
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.14,
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
"learning_rate": 0.00019615049043274205,
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 24
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15,
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
"learning_rate": 0.00019558514614057609,
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 25
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.15,
|
| 196 |
-
"grad_norm": 0.
|
| 197 |
"learning_rate": 0.00019498201317277828,
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 26
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.16,
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
"learning_rate": 0.00019434132997221345,
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 27
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.17,
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
"learning_rate": 0.0001936633498268728,
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 28
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.17,
|
| 217 |
-
"grad_norm": 0.
|
| 218 |
"learning_rate": 0.0001929483407697387,
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 29
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.18,
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
"learning_rate": 0.00019219658547282067,
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 30
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.18,
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
"learning_rate": 0.00019140838113540346,
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 31
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.19,
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
"learning_rate": 0.00019058403936655233,
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 32
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.2,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
"learning_rate": 0.00018972388606192125,
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 33
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.2,
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
"learning_rate": 0.0001888282612749132,
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 34
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.21,
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
"learning_rate": 0.00018789751908224338,
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 35
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.21,
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
"learning_rate": 0.00018693202744395827,
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 36
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.22,
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
"learning_rate": 0.00018593216805796612,
|
| 275 |
"loss": 0.5396,
|
| 276 |
"step": 37
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.23,
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
"learning_rate": 0.00018489833620913642,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 38
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.23,
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
"learning_rate": 0.00018383094061302766,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 39
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.24,
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
"learning_rate": 0.00018273040325430574,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.24,
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
"learning_rate": 0.00018159715921991612,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 41
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.25,
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
"learning_rate": 0.00018043165652707649,
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 42
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.25,
|
| 315 |
-
"eval_loss": 0.
|
| 316 |
-
"eval_runtime":
|
| 317 |
-
"eval_samples_per_second": 1.
|
| 318 |
-
"eval_steps_per_second": 0.
|
| 319 |
"step": 42
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 0.26,
|
| 323 |
-
"grad_norm": 0.
|
| 324 |
"learning_rate": 0.00017923435594615744,
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 43
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 0.26,
|
| 330 |
-
"grad_norm": 0.
|
| 331 |
"learning_rate": 0.00017800573081852122,
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 44
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 0.27,
|
| 337 |
-
"grad_norm": 0.
|
| 338 |
"learning_rate": 0.0001767462668693908,
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 45
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 0.27,
|
| 344 |
-
"grad_norm": 0.
|
| 345 |
"learning_rate": 0.00017545646201582303,
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 46
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 0.28,
|
| 351 |
-
"grad_norm": 0.
|
| 352 |
"learning_rate": 0.00017413682616986185,
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 47
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.29,
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
"learning_rate": 0.00017278788103694943,
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 48
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.29,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
"learning_rate": 0.000171410159909675,
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 49
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.3,
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
"learning_rate": 0.00017000420745694254,
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 50
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 0.3,
|
| 379 |
-
"grad_norm": 0.
|
| 380 |
"learning_rate": 0.00016857057950864132,
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 51
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 0.31,
|
| 386 |
-
"grad_norm": 0.
|
| 387 |
"learning_rate": 0.0001671098428359037,
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 52
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 0.32,
|
| 393 |
-
"grad_norm": 0.
|
| 394 |
"learning_rate": 0.00016562257492703757,
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 53
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.32,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
"learning_rate": 0.000164109363759222,
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 54
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.33,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 0.000162570807566056,
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 55
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.33,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
"learning_rate": 0.00016100751460105243,
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 56
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.34,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
"learning_rate": 0.00015942010289717105,
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 57
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 0.34,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
"learning_rate": 0.00015780920002248484,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 58
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.35,
|
| 435 |
-
"grad_norm": 0.
|
| 436 |
"learning_rate": 0.0001561754428320771,
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 59
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 0.36,
|
| 442 |
-
"grad_norm": 0.
|
| 443 |
"learning_rate": 0.00015451947721626676,
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 60
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 0.36,
|
| 449 |
-
"grad_norm": 0.
|
| 450 |
"learning_rate": 0.00015284195784526195,
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 61
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 0.37,
|
| 456 |
-
"grad_norm": 0.
|
| 457 |
"learning_rate": 0.00015114354791034225,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 62
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.37,
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
"learning_rate": 0.0001494249188616723,
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 63
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 0.38,
|
| 470 |
-
"grad_norm": 0.
|
| 471 |
"learning_rate": 0.00014768675014285062,
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 64
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.39,
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
"learning_rate": 0.00014592972892229778,
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 65
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 0.39,
|
| 484 |
-
"grad_norm": 0.
|
| 485 |
"learning_rate": 0.0001441545498215912,
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 66
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 0.4,
|
| 491 |
-
"grad_norm": 0.
|
| 492 |
"learning_rate": 0.00014236191464085286,
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 67
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 0.4,
|
| 498 |
-
"grad_norm": 0.
|
| 499 |
"learning_rate": 0.00014055253208129938,
|
| 500 |
-
"loss": 0.
|
| 501 |
"step": 68
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 0.41,
|
| 505 |
-
"grad_norm": 0.
|
| 506 |
"learning_rate": 0.00013872711746506413,
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 69
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 0.42,
|
| 512 |
-
"grad_norm": 0.
|
| 513 |
"learning_rate": 0.00013688639245240078,
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 70
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 0.42,
|
| 519 |
-
"grad_norm": 0.
|
| 520 |
"learning_rate": 0.00013503108475638244,
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 71
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 0.43,
|
| 526 |
-
"grad_norm": 0.
|
| 527 |
"learning_rate": 0.0001331619278552068,
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 72
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.43,
|
| 533 |
-
"grad_norm": 0.
|
| 534 |
"learning_rate": 0.00013127966070222274,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 73
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 0.44,
|
| 540 |
-
"grad_norm": 0.
|
| 541 |
"learning_rate": 0.00012938502743379212,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 74
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 0.45,
|
| 547 |
-
"grad_norm": 0.
|
| 548 |
"learning_rate": 0.00012747877707510252,
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 75
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 0.45,
|
| 554 |
-
"grad_norm": 0.
|
| 555 |
"learning_rate": 0.0001255616632440475,
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 76
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 0.46,
|
| 561 |
-
"grad_norm": 0.
|
| 562 |
"learning_rate": 0.0001236344438532905,
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 77
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.46,
|
| 568 |
-
"grad_norm": 0.
|
| 569 |
"learning_rate": 0.0001216978808106318,
|
| 570 |
"loss": 0.5091,
|
| 571 |
"step": 78
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 0.47,
|
| 575 |
-
"grad_norm": 0.
|
| 576 |
"learning_rate": 0.00011975273971779528,
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 79
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.48,
|
| 582 |
-
"grad_norm": 0.
|
| 583 |
"learning_rate": 0.00011779978956775506,
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 80
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.48,
|
| 589 |
-
"grad_norm": 0.
|
| 590 |
"learning_rate": 0.0001158398024407215,
|
| 591 |
-
"loss": 0.
|
| 592 |
"step": 81
|
| 593 |
},
|
| 594 |
{
|
| 595 |
"epoch": 0.49,
|
| 596 |
-
"grad_norm": 0.
|
| 597 |
"learning_rate": 0.00011387355319890685,
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 82
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 0.49,
|
| 603 |
-
"grad_norm": 0.
|
| 604 |
"learning_rate": 0.00011190181918019049,
|
| 605 |
-
"loss": 0.
|
| 606 |
"step": 83
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 0.5,
|
| 610 |
-
"grad_norm": 0.
|
| 611 |
"learning_rate": 0.00010992537989080618,
|
| 612 |
-
"loss": 0.
|
| 613 |
"step": 84
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 0.5,
|
| 617 |
-
"eval_loss": 0.
|
| 618 |
-
"eval_runtime":
|
| 619 |
-
"eval_samples_per_second": 1.
|
| 620 |
-
"eval_steps_per_second": 0.
|
| 621 |
"step": 84
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 0.51,
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
"learning_rate": 0.00010794501669717145,
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 85
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 0.51,
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
"learning_rate": 0.00010596151251698199,
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 86
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 0.52,
|
| 639 |
-
"grad_norm": 0.
|
| 640 |
"learning_rate": 0.0001039756515096926,
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 87
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 0.52,
|
| 646 |
-
"grad_norm": 0.
|
| 647 |
"learning_rate": 0.00010198821876650701,
|
| 648 |
-
"loss": 0.
|
| 649 |
"step": 88
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.53,
|
| 653 |
-
"grad_norm": 0.
|
| 654 |
"learning_rate": 0.0001,
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 89
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 0.53,
|
| 660 |
-
"grad_norm": 0.
|
| 661 |
"learning_rate": 9.801178123349298e-05,
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 90
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 0.54,
|
| 667 |
-
"grad_norm": 0.
|
| 668 |
"learning_rate": 9.602434849030745e-05,
|
| 669 |
-
"loss": 0.
|
| 670 |
"step": 91
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 0.55,
|
| 674 |
-
"grad_norm": 0.
|
| 675 |
"learning_rate": 9.403848748301802e-05,
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 92
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 0.55,
|
| 681 |
-
"grad_norm": 0.
|
| 682 |
"learning_rate": 9.205498330282856e-05,
|
| 683 |
-
"loss": 0.
|
| 684 |
"step": 93
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.56,
|
| 688 |
-
"grad_norm": 0.
|
| 689 |
"learning_rate": 9.007462010919386e-05,
|
| 690 |
-
"loss": 0.
|
| 691 |
"step": 94
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 0.56,
|
| 695 |
-
"grad_norm": 0.
|
| 696 |
"learning_rate": 8.809818081980953e-05,
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 95
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 0.57,
|
| 702 |
-
"grad_norm": 0.
|
| 703 |
"learning_rate": 8.612644680109319e-05,
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 96
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.58,
|
| 709 |
-
"grad_norm": 0.
|
| 710 |
"learning_rate": 8.416019755927851e-05,
|
| 711 |
-
"loss": 0.
|
| 712 |
"step": 97
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 0.58,
|
| 716 |
-
"grad_norm": 0.
|
| 717 |
"learning_rate": 8.2200210432245e-05,
|
| 718 |
-
"loss": 0.
|
| 719 |
"step": 98
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 0.59,
|
| 723 |
-
"grad_norm": 0.
|
| 724 |
"learning_rate": 8.024726028220474e-05,
|
| 725 |
-
"loss": 0.
|
| 726 |
"step": 99
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 0.59,
|
| 730 |
-
"grad_norm": 0.
|
| 731 |
"learning_rate": 7.83021191893682e-05,
|
| 732 |
-
"loss": 0.
|
| 733 |
"step": 100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 0.6,
|
| 737 |
-
"grad_norm": 0.
|
| 738 |
"learning_rate": 7.636555614670953e-05,
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 101
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.61,
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
"learning_rate": 7.443833675595255e-05,
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 102
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.61,
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
"learning_rate": 7.252122292489747e-05,
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 103
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.62,
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
"learning_rate": 7.061497256620793e-05,
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 104
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.62,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 6.87203392977773e-05,
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 105
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.63,
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
"learning_rate": 6.683807214479323e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 106
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.64,
|
| 779 |
-
"grad_norm": 0.
|
| 780 |
"learning_rate": 6.496891524361757e-05,
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 107
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 0.64,
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
"learning_rate": 6.311360754759923e-05,
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 108
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.65,
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
"learning_rate": 6.127288253493591e-05,
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 109
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 0.65,
|
| 800 |
-
"grad_norm": 0.
|
| 801 |
"learning_rate": 5.9447467918700614e-05,
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 110
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 0.66,
|
| 807 |
-
"grad_norm": 0.
|
| 808 |
"learning_rate": 5.763808535914723e-05,
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 111
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 0.67,
|
| 814 |
-
"grad_norm": 0.
|
| 815 |
"learning_rate": 5.584545017840885e-05,
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 112
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.67,
|
| 821 |
-
"grad_norm": 0.
|
| 822 |
"learning_rate": 5.407027107770219e-05,
|
| 823 |
-
"loss": 0.
|
| 824 |
"step": 113
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 0.68,
|
| 828 |
-
"grad_norm": 0.
|
| 829 |
"learning_rate": 5.2313249857149414e-05,
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 114
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 0.68,
|
| 835 |
-
"grad_norm": 0.
|
| 836 |
"learning_rate": 5.0575081138327715e-05,
|
| 837 |
-
"loss": 0.
|
| 838 |
"step": 115
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 0.69,
|
| 842 |
-
"grad_norm": 0.
|
| 843 |
"learning_rate": 4.885645208965779e-05,
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 116
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 0.7,
|
| 849 |
-
"grad_norm": 0.
|
| 850 |
"learning_rate": 4.715804215473809e-05,
|
| 851 |
-
"loss": 0.
|
| 852 |
"step": 117
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 0.7,
|
| 856 |
-
"grad_norm": 0.
|
| 857 |
"learning_rate": 4.548052278373327e-05,
|
| 858 |
-
"loss": 0.
|
| 859 |
"step": 118
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 0.71,
|
| 863 |
-
"grad_norm": 0.
|
| 864 |
"learning_rate": 4.382455716792291e-05,
|
| 865 |
-
"loss": 0.
|
| 866 |
"step": 119
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 0.71,
|
| 870 |
-
"grad_norm": 0.
|
| 871 |
"learning_rate": 4.219079997751515e-05,
|
| 872 |
"loss": 0.4954,
|
| 873 |
"step": 120
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.72,
|
| 877 |
-
"grad_norm": 0.
|
| 878 |
"learning_rate": 4.0579897102828966e-05,
|
| 879 |
-
"loss": 0.
|
| 880 |
"step": 121
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 0.73,
|
| 884 |
-
"grad_norm": 0.
|
| 885 |
"learning_rate": 3.899248539894757e-05,
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 122
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 0.73,
|
| 891 |
-
"grad_norm": 0.
|
| 892 |
"learning_rate": 3.7429192433944014e-05,
|
| 893 |
-
"loss": 0.
|
| 894 |
"step": 123
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.74,
|
| 898 |
-
"grad_norm": 0.
|
| 899 |
"learning_rate": 3.589063624077802e-05,
|
| 900 |
-
"loss": 0.
|
| 901 |
"step": 124
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.74,
|
| 905 |
-
"grad_norm": 0.
|
| 906 |
"learning_rate": 3.4377425072962465e-05,
|
| 907 |
-
"loss": 0.
|
| 908 |
"step": 125
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 0.75,
|
| 912 |
-
"grad_norm": 0.
|
| 913 |
"learning_rate": 3.289015716409631e-05,
|
| 914 |
-
"loss": 0.
|
| 915 |
"step": 126
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 0.75,
|
| 919 |
-
"eval_loss": 0.
|
| 920 |
-
"eval_runtime": 155.
|
| 921 |
-
"eval_samples_per_second": 1.
|
| 922 |
-
"eval_steps_per_second": 0.
|
| 923 |
"step": 126
|
| 924 |
}
|
| 925 |
],
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.29289755909939047,
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
"loss": 0.6274,
|
| 16 |
"step": 1
|
|
|
|
| 18 |
{
|
| 19 |
"epoch": 0.01,
|
| 20 |
"eval_loss": 1.0297596454620361,
|
| 21 |
+
"eval_runtime": 153.0715,
|
| 22 |
+
"eval_samples_per_second": 1.627,
|
| 23 |
+
"eval_steps_per_second": 0.817,
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.01,
|
| 28 |
+
"grad_norm": 0.27956410941469395,
|
| 29 |
"learning_rate": 4e-05,
|
| 30 |
"loss": 0.6362,
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.02,
|
| 35 |
+
"grad_norm": 0.3009590515152092,
|
| 36 |
"learning_rate": 6e-05,
|
| 37 |
"loss": 0.6299,
|
| 38 |
"step": 3
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.02,
|
| 42 |
+
"grad_norm": 0.34354546270515235,
|
| 43 |
"learning_rate": 8e-05,
|
| 44 |
+
"loss": 0.6395,
|
| 45 |
"step": 4
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.03,
|
| 49 |
+
"grad_norm": 0.2470961998205002,
|
| 50 |
"learning_rate": 0.0001,
|
| 51 |
+
"loss": 0.6068,
|
| 52 |
"step": 5
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.04,
|
| 56 |
+
"grad_norm": 0.2071993792912813,
|
| 57 |
"learning_rate": 0.00012,
|
| 58 |
+
"loss": 0.5993,
|
| 59 |
"step": 6
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.04,
|
| 63 |
+
"grad_norm": 0.20237019487582247,
|
| 64 |
"learning_rate": 0.00014,
|
| 65 |
+
"loss": 0.6293,
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.05,
|
| 70 |
+
"grad_norm": 0.13810925455451734,
|
| 71 |
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 0.5101,
|
| 73 |
"step": 8
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.05,
|
| 77 |
+
"grad_norm": 0.22762469698117493,
|
| 78 |
"learning_rate": 0.00018,
|
| 79 |
+
"loss": 0.5527,
|
| 80 |
"step": 9
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.06,
|
| 84 |
+
"grad_norm": 0.15718586910417978,
|
| 85 |
"learning_rate": 0.0002,
|
| 86 |
+
"loss": 0.5437,
|
| 87 |
"step": 10
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.07,
|
| 91 |
+
"grad_norm": 0.13825435646308676,
|
| 92 |
"learning_rate": 0.00019998023297700658,
|
| 93 |
+
"loss": 0.5858,
|
| 94 |
"step": 11
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.07,
|
| 98 |
+
"grad_norm": 0.15072585803075744,
|
| 99 |
"learning_rate": 0.00019992093972273018,
|
| 100 |
+
"loss": 0.554,
|
| 101 |
"step": 12
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.08,
|
| 105 |
+
"grad_norm": 0.13616736806741778,
|
| 106 |
"learning_rate": 0.00019982214367819328,
|
| 107 |
+
"loss": 0.6199,
|
| 108 |
"step": 13
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.08,
|
| 112 |
+
"grad_norm": 0.15841398198140327,
|
| 113 |
"learning_rate": 0.0001996838839014696,
|
| 114 |
"loss": 0.5495,
|
| 115 |
"step": 14
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.09,
|
| 119 |
+
"grad_norm": 0.16548492090565414,
|
| 120 |
"learning_rate": 0.00019950621505224273,
|
| 121 |
+
"loss": 0.5043,
|
| 122 |
"step": 15
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.1,
|
| 126 |
+
"grad_norm": 0.16965923620151113,
|
| 127 |
"learning_rate": 0.00019928920737019733,
|
| 128 |
"loss": 0.5083,
|
| 129 |
"step": 16
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.1,
|
| 133 |
+
"grad_norm": 0.1853247768877015,
|
| 134 |
"learning_rate": 0.0001990329466472502,
|
| 135 |
+
"loss": 0.6318,
|
| 136 |
"step": 17
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.11,
|
| 140 |
+
"grad_norm": 0.1824469971767483,
|
| 141 |
"learning_rate": 0.00019873753419363336,
|
| 142 |
+
"loss": 0.4809,
|
| 143 |
"step": 18
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.11,
|
| 147 |
+
"grad_norm": 0.13288488939703025,
|
| 148 |
"learning_rate": 0.00019840308679784207,
|
| 149 |
+
"loss": 0.4974,
|
| 150 |
"step": 19
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.12,
|
| 154 |
+
"grad_norm": 0.18507602579352214,
|
| 155 |
"learning_rate": 0.00019802973668046363,
|
| 156 |
+
"loss": 0.5288,
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.12,
|
| 161 |
+
"grad_norm": 0.13591012925799634,
|
| 162 |
"learning_rate": 0.0001976176314419051,
|
| 163 |
+
"loss": 0.5291,
|
| 164 |
"step": 21
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.13,
|
| 168 |
+
"grad_norm": 0.13704233024715257,
|
| 169 |
"learning_rate": 0.000197166934004041,
|
| 170 |
"loss": 0.4819,
|
| 171 |
"step": 22
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.14,
|
| 175 |
+
"grad_norm": 0.15362008062157553,
|
| 176 |
"learning_rate": 0.00019667782254580374,
|
| 177 |
+
"loss": 0.5408,
|
| 178 |
"step": 23
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.14,
|
| 182 |
+
"grad_norm": 0.11171032733778614,
|
| 183 |
"learning_rate": 0.00019615049043274205,
|
| 184 |
+
"loss": 0.5101,
|
| 185 |
"step": 24
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15,
|
| 189 |
+
"grad_norm": 0.11449051252849428,
|
| 190 |
"learning_rate": 0.00019558514614057609,
|
| 191 |
+
"loss": 0.5209,
|
| 192 |
"step": 25
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.15,
|
| 196 |
+
"grad_norm": 0.11611407915744347,
|
| 197 |
"learning_rate": 0.00019498201317277828,
|
| 198 |
+
"loss": 0.5005,
|
| 199 |
"step": 26
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.16,
|
| 203 |
+
"grad_norm": 0.12735633641627706,
|
| 204 |
"learning_rate": 0.00019434132997221345,
|
| 205 |
+
"loss": 0.4741,
|
| 206 |
"step": 27
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.17,
|
| 210 |
+
"grad_norm": 0.11871518327376328,
|
| 211 |
"learning_rate": 0.0001936633498268728,
|
| 212 |
+
"loss": 0.5213,
|
| 213 |
"step": 28
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.17,
|
| 217 |
+
"grad_norm": 0.11403376465541806,
|
| 218 |
"learning_rate": 0.0001929483407697387,
|
| 219 |
+
"loss": 0.4842,
|
| 220 |
"step": 29
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.18,
|
| 224 |
+
"grad_norm": 0.11675155934307391,
|
| 225 |
"learning_rate": 0.00019219658547282067,
|
| 226 |
+
"loss": 0.4825,
|
| 227 |
"step": 30
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.18,
|
| 231 |
+
"grad_norm": 0.10789338581384152,
|
| 232 |
"learning_rate": 0.00019140838113540346,
|
| 233 |
+
"loss": 0.4866,
|
| 234 |
"step": 31
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.19,
|
| 238 |
+
"grad_norm": 0.14336837303756964,
|
| 239 |
"learning_rate": 0.00019058403936655233,
|
| 240 |
+
"loss": 0.5325,
|
| 241 |
"step": 32
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.2,
|
| 245 |
+
"grad_norm": 0.10401694793599091,
|
| 246 |
"learning_rate": 0.00018972388606192125,
|
| 247 |
+
"loss": 0.4292,
|
| 248 |
"step": 33
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.2,
|
| 252 |
+
"grad_norm": 0.10614245315138679,
|
| 253 |
"learning_rate": 0.0001888282612749132,
|
| 254 |
+
"loss": 0.4638,
|
| 255 |
"step": 34
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.21,
|
| 259 |
+
"grad_norm": 0.1250509143492961,
|
| 260 |
"learning_rate": 0.00018789751908224338,
|
| 261 |
+
"loss": 0.4963,
|
| 262 |
"step": 35
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.21,
|
| 266 |
+
"grad_norm": 0.12668831423294083,
|
| 267 |
"learning_rate": 0.00018693202744395827,
|
| 268 |
+
"loss": 0.5043,
|
| 269 |
"step": 36
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.22,
|
| 273 |
+
"grad_norm": 0.11594924793976216,
|
| 274 |
"learning_rate": 0.00018593216805796612,
|
| 275 |
"loss": 0.5396,
|
| 276 |
"step": 37
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.23,
|
| 280 |
+
"grad_norm": 0.12365894749489846,
|
| 281 |
"learning_rate": 0.00018489833620913642,
|
| 282 |
+
"loss": 0.4899,
|
| 283 |
"step": 38
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.23,
|
| 287 |
+
"grad_norm": 0.13164318970805183,
|
| 288 |
"learning_rate": 0.00018383094061302766,
|
| 289 |
+
"loss": 0.5065,
|
| 290 |
"step": 39
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.24,
|
| 294 |
+
"grad_norm": 0.10587870431925453,
|
| 295 |
"learning_rate": 0.00018273040325430574,
|
| 296 |
+
"loss": 0.4805,
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.24,
|
| 301 |
+
"grad_norm": 0.12134503625595237,
|
| 302 |
"learning_rate": 0.00018159715921991612,
|
| 303 |
+
"loss": 0.5103,
|
| 304 |
"step": 41
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.25,
|
| 308 |
+
"grad_norm": 0.10855490758268896,
|
| 309 |
"learning_rate": 0.00018043165652707649,
|
| 310 |
+
"loss": 0.44,
|
| 311 |
"step": 42
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.25,
|
| 315 |
+
"eval_loss": 0.9770342707633972,
|
| 316 |
+
"eval_runtime": 154.3933,
|
| 317 |
+
"eval_samples_per_second": 1.613,
|
| 318 |
+
"eval_steps_per_second": 0.81,
|
| 319 |
"step": 42
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 0.26,
|
| 323 |
+
"grad_norm": 0.11753509152196469,
|
| 324 |
"learning_rate": 0.00017923435594615744,
|
| 325 |
+
"loss": 0.4819,
|
| 326 |
"step": 43
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 0.26,
|
| 330 |
+
"grad_norm": 0.13774603727779988,
|
| 331 |
"learning_rate": 0.00017800573081852122,
|
| 332 |
+
"loss": 0.5451,
|
| 333 |
"step": 44
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 0.27,
|
| 337 |
+
"grad_norm": 0.11994636996852912,
|
| 338 |
"learning_rate": 0.0001767462668693908,
|
| 339 |
+
"loss": 0.5079,
|
| 340 |
"step": 45
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 0.27,
|
| 344 |
+
"grad_norm": 0.11803018063108017,
|
| 345 |
"learning_rate": 0.00017545646201582303,
|
| 346 |
+
"loss": 0.5183,
|
| 347 |
"step": 46
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 0.28,
|
| 351 |
+
"grad_norm": 0.12122026879022209,
|
| 352 |
"learning_rate": 0.00017413682616986185,
|
| 353 |
+
"loss": 0.4692,
|
| 354 |
"step": 47
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.29,
|
| 358 |
+
"grad_norm": 0.12840154129375927,
|
| 359 |
"learning_rate": 0.00017278788103694943,
|
| 360 |
+
"loss": 0.4538,
|
| 361 |
"step": 48
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.29,
|
| 365 |
+
"grad_norm": 0.10754191454075242,
|
| 366 |
"learning_rate": 0.000171410159909675,
|
| 367 |
+
"loss": 0.4745,
|
| 368 |
"step": 49
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.3,
|
| 372 |
+
"grad_norm": 0.10980392972154758,
|
| 373 |
"learning_rate": 0.00017000420745694254,
|
| 374 |
+
"loss": 0.5077,
|
| 375 |
"step": 50
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 0.3,
|
| 379 |
+
"grad_norm": 0.1018314102740997,
|
| 380 |
"learning_rate": 0.00016857057950864132,
|
| 381 |
+
"loss": 0.5077,
|
| 382 |
"step": 51
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 0.31,
|
| 386 |
+
"grad_norm": 0.10572512690181787,
|
| 387 |
"learning_rate": 0.0001671098428359037,
|
| 388 |
+
"loss": 0.4637,
|
| 389 |
"step": 52
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 0.32,
|
| 393 |
+
"grad_norm": 0.13106813432864392,
|
| 394 |
"learning_rate": 0.00016562257492703757,
|
| 395 |
+
"loss": 0.4718,
|
| 396 |
"step": 53
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.32,
|
| 400 |
+
"grad_norm": 0.11024484284605006,
|
| 401 |
"learning_rate": 0.000164109363759222,
|
| 402 |
+
"loss": 0.5115,
|
| 403 |
"step": 54
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.33,
|
| 407 |
+
"grad_norm": 0.12388815222110366,
|
| 408 |
"learning_rate": 0.000162570807566056,
|
| 409 |
+
"loss": 0.4756,
|
| 410 |
"step": 55
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.33,
|
| 414 |
+
"grad_norm": 0.12702188830349206,
|
| 415 |
"learning_rate": 0.00016100751460105243,
|
| 416 |
+
"loss": 0.4881,
|
| 417 |
"step": 56
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.34,
|
| 421 |
+
"grad_norm": 0.11086308659932648,
|
| 422 |
"learning_rate": 0.00015942010289717105,
|
| 423 |
+
"loss": 0.4701,
|
| 424 |
"step": 57
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 0.34,
|
| 428 |
+
"grad_norm": 0.11517890306226485,
|
| 429 |
"learning_rate": 0.00015780920002248484,
|
| 430 |
+
"loss": 0.4835,
|
| 431 |
"step": 58
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.35,
|
| 435 |
+
"grad_norm": 0.1253541013817757,
|
| 436 |
"learning_rate": 0.0001561754428320771,
|
| 437 |
+
"loss": 0.5147,
|
| 438 |
"step": 59
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 0.36,
|
| 442 |
+
"grad_norm": 0.11139862438389284,
|
| 443 |
"learning_rate": 0.00015451947721626676,
|
| 444 |
+
"loss": 0.4552,
|
| 445 |
"step": 60
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 0.36,
|
| 449 |
+
"grad_norm": 0.1152466912880715,
|
| 450 |
"learning_rate": 0.00015284195784526195,
|
| 451 |
+
"loss": 0.5053,
|
| 452 |
"step": 61
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 0.37,
|
| 456 |
+
"grad_norm": 0.11719477948785388,
|
| 457 |
"learning_rate": 0.00015114354791034225,
|
| 458 |
+
"loss": 0.5079,
|
| 459 |
"step": 62
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.37,
|
| 463 |
+
"grad_norm": 0.12730957754598396,
|
| 464 |
"learning_rate": 0.0001494249188616723,
|
| 465 |
+
"loss": 0.58,
|
| 466 |
"step": 63
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 0.38,
|
| 470 |
+
"grad_norm": 0.11907511614241814,
|
| 471 |
"learning_rate": 0.00014768675014285062,
|
| 472 |
+
"loss": 0.4579,
|
| 473 |
"step": 64
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.39,
|
| 477 |
+
"grad_norm": 0.13613222195950112,
|
| 478 |
"learning_rate": 0.00014592972892229778,
|
| 479 |
+
"loss": 0.4965,
|
| 480 |
"step": 65
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 0.39,
|
| 484 |
+
"grad_norm": 0.11610538032286319,
|
| 485 |
"learning_rate": 0.0001441545498215912,
|
| 486 |
+
"loss": 0.4467,
|
| 487 |
"step": 66
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 0.4,
|
| 491 |
+
"grad_norm": 0.1046724680527729,
|
| 492 |
"learning_rate": 0.00014236191464085286,
|
| 493 |
+
"loss": 0.4463,
|
| 494 |
"step": 67
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 0.4,
|
| 498 |
+
"grad_norm": 0.11631551012538931,
|
| 499 |
"learning_rate": 0.00014055253208129938,
|
| 500 |
+
"loss": 0.5304,
|
| 501 |
"step": 68
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 0.41,
|
| 505 |
+
"grad_norm": 0.11792788666231199,
|
| 506 |
"learning_rate": 0.00013872711746506413,
|
| 507 |
+
"loss": 0.453,
|
| 508 |
"step": 69
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 0.42,
|
| 512 |
+
"grad_norm": 0.11566019183159004,
|
| 513 |
"learning_rate": 0.00013688639245240078,
|
| 514 |
+
"loss": 0.5192,
|
| 515 |
"step": 70
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 0.42,
|
| 519 |
+
"grad_norm": 0.12967661214418025,
|
| 520 |
"learning_rate": 0.00013503108475638244,
|
| 521 |
+
"loss": 0.5625,
|
| 522 |
"step": 71
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 0.43,
|
| 526 |
+
"grad_norm": 0.10928605499181634,
|
| 527 |
"learning_rate": 0.0001331619278552068,
|
| 528 |
+
"loss": 0.4861,
|
| 529 |
"step": 72
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.43,
|
| 533 |
+
"grad_norm": 0.12485741141890881,
|
| 534 |
"learning_rate": 0.00013127966070222274,
|
| 535 |
+
"loss": 0.4782,
|
| 536 |
"step": 73
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 0.44,
|
| 540 |
+
"grad_norm": 0.1271574869759,
|
| 541 |
"learning_rate": 0.00012938502743379212,
|
| 542 |
+
"loss": 0.4819,
|
| 543 |
"step": 74
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 0.45,
|
| 547 |
+
"grad_norm": 0.14528941719728583,
|
| 548 |
"learning_rate": 0.00012747877707510252,
|
| 549 |
+
"loss": 0.5132,
|
| 550 |
"step": 75
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 0.45,
|
| 554 |
+
"grad_norm": 0.12760881412243183,
|
| 555 |
"learning_rate": 0.0001255616632440475,
|
| 556 |
+
"loss": 0.4817,
|
| 557 |
"step": 76
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 0.46,
|
| 561 |
+
"grad_norm": 0.1239773484446177,
|
| 562 |
"learning_rate": 0.0001236344438532905,
|
| 563 |
+
"loss": 0.5045,
|
| 564 |
"step": 77
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.46,
|
| 568 |
+
"grad_norm": 0.1307497276864945,
|
| 569 |
"learning_rate": 0.0001216978808106318,
|
| 570 |
"loss": 0.5091,
|
| 571 |
"step": 78
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 0.47,
|
| 575 |
+
"grad_norm": 0.1355749101256534,
|
| 576 |
"learning_rate": 0.00011975273971779528,
|
| 577 |
+
"loss": 0.5145,
|
| 578 |
"step": 79
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.48,
|
| 582 |
+
"grad_norm": 0.12394948316350356,
|
| 583 |
"learning_rate": 0.00011779978956775506,
|
| 584 |
+
"loss": 0.5057,
|
| 585 |
"step": 80
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.48,
|
| 589 |
+
"grad_norm": 0.13424789799433426,
|
| 590 |
"learning_rate": 0.0001158398024407215,
|
| 591 |
+
"loss": 0.5058,
|
| 592 |
"step": 81
|
| 593 |
},
|
| 594 |
{
|
| 595 |
"epoch": 0.49,
|
| 596 |
+
"grad_norm": 0.13054427599188898,
|
| 597 |
"learning_rate": 0.00011387355319890685,
|
| 598 |
+
"loss": 0.4683,
|
| 599 |
"step": 82
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 0.49,
|
| 603 |
+
"grad_norm": 0.11996704114604598,
|
| 604 |
"learning_rate": 0.00011190181918019049,
|
| 605 |
+
"loss": 0.4748,
|
| 606 |
"step": 83
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 0.5,
|
| 610 |
+
"grad_norm": 0.13071208464837492,
|
| 611 |
"learning_rate": 0.00010992537989080618,
|
| 612 |
+
"loss": 0.4406,
|
| 613 |
"step": 84
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 0.5,
|
| 617 |
+
"eval_loss": 0.9652944803237915,
|
| 618 |
+
"eval_runtime": 153.5377,
|
| 619 |
+
"eval_samples_per_second": 1.622,
|
| 620 |
+
"eval_steps_per_second": 0.814,
|
| 621 |
"step": 84
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 0.51,
|
| 625 |
+
"grad_norm": 0.11393059638442392,
|
| 626 |
"learning_rate": 0.00010794501669717145,
|
| 627 |
+
"loss": 0.4877,
|
| 628 |
"step": 85
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 0.51,
|
| 632 |
+
"grad_norm": 0.11364636282385192,
|
| 633 |
"learning_rate": 0.00010596151251698199,
|
| 634 |
+
"loss": 0.4597,
|
| 635 |
"step": 86
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 0.52,
|
| 639 |
+
"grad_norm": 0.12548992883216656,
|
| 640 |
"learning_rate": 0.0001039756515096926,
|
| 641 |
+
"loss": 0.4685,
|
| 642 |
"step": 87
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 0.52,
|
| 646 |
+
"grad_norm": 0.12825123946992636,
|
| 647 |
"learning_rate": 0.00010198821876650701,
|
| 648 |
+
"loss": 0.4924,
|
| 649 |
"step": 88
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.53,
|
| 653 |
+
"grad_norm": 0.12264617890724591,
|
| 654 |
"learning_rate": 0.0001,
|
| 655 |
+
"loss": 0.4678,
|
| 656 |
"step": 89
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 0.53,
|
| 660 |
+
"grad_norm": 0.1238990144553216,
|
| 661 |
"learning_rate": 9.801178123349298e-05,
|
| 662 |
+
"loss": 0.4854,
|
| 663 |
"step": 90
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 0.54,
|
| 667 |
+
"grad_norm": 0.13533454171482565,
|
| 668 |
"learning_rate": 9.602434849030745e-05,
|
| 669 |
+
"loss": 0.4784,
|
| 670 |
"step": 91
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 0.55,
|
| 674 |
+
"grad_norm": 0.13623990945679543,
|
| 675 |
"learning_rate": 9.403848748301802e-05,
|
| 676 |
+
"loss": 0.5322,
|
| 677 |
"step": 92
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 0.55,
|
| 681 |
+
"grad_norm": 0.16411366119133766,
|
| 682 |
"learning_rate": 9.205498330282856e-05,
|
| 683 |
+
"loss": 0.5258,
|
| 684 |
"step": 93
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.56,
|
| 688 |
+
"grad_norm": 0.11646811294381437,
|
| 689 |
"learning_rate": 9.007462010919386e-05,
|
| 690 |
+
"loss": 0.4599,
|
| 691 |
"step": 94
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 0.56,
|
| 695 |
+
"grad_norm": 0.12610649830308363,
|
| 696 |
"learning_rate": 8.809818081980953e-05,
|
| 697 |
+
"loss": 0.4891,
|
| 698 |
"step": 95
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 0.57,
|
| 702 |
+
"grad_norm": 0.12324596375061997,
|
| 703 |
"learning_rate": 8.612644680109319e-05,
|
| 704 |
+
"loss": 0.4771,
|
| 705 |
"step": 96
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.58,
|
| 709 |
+
"grad_norm": 0.12390720748290898,
|
| 710 |
"learning_rate": 8.416019755927851e-05,
|
| 711 |
+
"loss": 0.4814,
|
| 712 |
"step": 97
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 0.58,
|
| 716 |
+
"grad_norm": 0.11156155557793686,
|
| 717 |
"learning_rate": 8.2200210432245e-05,
|
| 718 |
+
"loss": 0.5041,
|
| 719 |
"step": 98
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 0.59,
|
| 723 |
+
"grad_norm": 0.11235071151397323,
|
| 724 |
"learning_rate": 8.024726028220474e-05,
|
| 725 |
+
"loss": 0.5023,
|
| 726 |
"step": 99
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 0.59,
|
| 730 |
+
"grad_norm": 0.12382934807374943,
|
| 731 |
"learning_rate": 7.83021191893682e-05,
|
| 732 |
+
"loss": 0.4917,
|
| 733 |
"step": 100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 0.6,
|
| 737 |
+
"grad_norm": 0.13051565063971357,
|
| 738 |
"learning_rate": 7.636555614670953e-05,
|
| 739 |
+
"loss": 0.4567,
|
| 740 |
"step": 101
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.61,
|
| 744 |
+
"grad_norm": 0.11975246976802223,
|
| 745 |
"learning_rate": 7.443833675595255e-05,
|
| 746 |
+
"loss": 0.4584,
|
| 747 |
"step": 102
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.61,
|
| 751 |
+
"grad_norm": 0.1390325061190883,
|
| 752 |
"learning_rate": 7.252122292489747e-05,
|
| 753 |
+
"loss": 0.4853,
|
| 754 |
"step": 103
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.62,
|
| 758 |
+
"grad_norm": 0.13092284300351015,
|
| 759 |
"learning_rate": 7.061497256620793e-05,
|
| 760 |
+
"loss": 0.4623,
|
| 761 |
"step": 104
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.62,
|
| 765 |
+
"grad_norm": 0.1291139691922856,
|
| 766 |
"learning_rate": 6.87203392977773e-05,
|
| 767 |
+
"loss": 0.4826,
|
| 768 |
"step": 105
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.63,
|
| 772 |
+
"grad_norm": 0.12630345566649634,
|
| 773 |
"learning_rate": 6.683807214479323e-05,
|
| 774 |
+
"loss": 0.4597,
|
| 775 |
"step": 106
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.64,
|
| 779 |
+
"grad_norm": 0.13061642512554117,
|
| 780 |
"learning_rate": 6.496891524361757e-05,
|
| 781 |
+
"loss": 0.4415,
|
| 782 |
"step": 107
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 0.64,
|
| 786 |
+
"grad_norm": 0.11414947835097713,
|
| 787 |
"learning_rate": 6.311360754759923e-05,
|
| 788 |
+
"loss": 0.4011,
|
| 789 |
"step": 108
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.65,
|
| 793 |
+
"grad_norm": 0.11453014913188615,
|
| 794 |
"learning_rate": 6.127288253493591e-05,
|
| 795 |
+
"loss": 0.5118,
|
| 796 |
"step": 109
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 0.65,
|
| 800 |
+
"grad_norm": 0.15000233747727326,
|
| 801 |
"learning_rate": 5.9447467918700614e-05,
|
| 802 |
+
"loss": 0.482,
|
| 803 |
"step": 110
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 0.66,
|
| 807 |
+
"grad_norm": 0.14402898946913723,
|
| 808 |
"learning_rate": 5.763808535914723e-05,
|
| 809 |
+
"loss": 0.4881,
|
| 810 |
"step": 111
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 0.67,
|
| 814 |
+
"grad_norm": 0.1371953423024685,
|
| 815 |
"learning_rate": 5.584545017840885e-05,
|
| 816 |
+
"loss": 0.5178,
|
| 817 |
"step": 112
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.67,
|
| 821 |
+
"grad_norm": 0.14105581149485616,
|
| 822 |
"learning_rate": 5.407027107770219e-05,
|
| 823 |
+
"loss": 0.5584,
|
| 824 |
"step": 113
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 0.68,
|
| 828 |
+
"grad_norm": 0.13646035299068404,
|
| 829 |
"learning_rate": 5.2313249857149414e-05,
|
| 830 |
+
"loss": 0.4535,
|
| 831 |
"step": 114
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 0.68,
|
| 835 |
+
"grad_norm": 0.14741390311838884,
|
| 836 |
"learning_rate": 5.0575081138327715e-05,
|
| 837 |
+
"loss": 0.5,
|
| 838 |
"step": 115
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 0.69,
|
| 842 |
+
"grad_norm": 0.12313483515418376,
|
| 843 |
"learning_rate": 4.885645208965779e-05,
|
| 844 |
+
"loss": 0.4982,
|
| 845 |
"step": 116
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 0.7,
|
| 849 |
+
"grad_norm": 0.13192008123654636,
|
| 850 |
"learning_rate": 4.715804215473809e-05,
|
| 851 |
+
"loss": 0.4698,
|
| 852 |
"step": 117
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 0.7,
|
| 856 |
+
"grad_norm": 0.11197839688620534,
|
| 857 |
"learning_rate": 4.548052278373327e-05,
|
| 858 |
+
"loss": 0.4737,
|
| 859 |
"step": 118
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 0.71,
|
| 863 |
+
"grad_norm": 0.12574619847347493,
|
| 864 |
"learning_rate": 4.382455716792291e-05,
|
| 865 |
+
"loss": 0.472,
|
| 866 |
"step": 119
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 0.71,
|
| 870 |
+
"grad_norm": 0.11785384512597769,
|
| 871 |
"learning_rate": 4.219079997751515e-05,
|
| 872 |
"loss": 0.4954,
|
| 873 |
"step": 120
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.72,
|
| 877 |
+
"grad_norm": 0.1371137498771671,
|
| 878 |
"learning_rate": 4.0579897102828966e-05,
|
| 879 |
+
"loss": 0.4645,
|
| 880 |
"step": 121
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 0.73,
|
| 884 |
+
"grad_norm": 0.1562115085397725,
|
| 885 |
"learning_rate": 3.899248539894757e-05,
|
| 886 |
+
"loss": 0.4798,
|
| 887 |
"step": 122
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 0.73,
|
| 891 |
+
"grad_norm": 0.14251919403809987,
|
| 892 |
"learning_rate": 3.7429192433944014e-05,
|
| 893 |
+
"loss": 0.4794,
|
| 894 |
"step": 123
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.74,
|
| 898 |
+
"grad_norm": 0.13567896269328303,
|
| 899 |
"learning_rate": 3.589063624077802e-05,
|
| 900 |
+
"loss": 0.4441,
|
| 901 |
"step": 124
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.74,
|
| 905 |
+
"grad_norm": 0.14154096161651117,
|
| 906 |
"learning_rate": 3.4377425072962465e-05,
|
| 907 |
+
"loss": 0.4583,
|
| 908 |
"step": 125
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 0.75,
|
| 912 |
+
"grad_norm": 0.12685010104316322,
|
| 913 |
"learning_rate": 3.289015716409631e-05,
|
| 914 |
+
"loss": 0.4445,
|
| 915 |
"step": 126
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 0.75,
|
| 919 |
+
"eval_loss": 0.9644750952720642,
|
| 920 |
+
"eval_runtime": 155.1633,
|
| 921 |
+
"eval_samples_per_second": 1.605,
|
| 922 |
+
"eval_steps_per_second": 0.806,
|
| 923 |
"step": 126
|
| 924 |
}
|
| 925 |
],
|
checkpoint-126/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6968
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3baf97ea9bcc6b73a05f59bedb019ebf1656742a3d0668caac6a61c4da440046
|
| 3 |
size 6968
|
checkpoint-168/adapter_config.json
CHANGED
|
@@ -20,17 +20,17 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"
|
| 24 |
-
"o_proj",
|
| 25 |
-
"up_proj",
|
| 26 |
"v_proj",
|
| 27 |
-
"down_proj",
|
| 28 |
"dt_proj",
|
| 29 |
-
"
|
| 30 |
-
"q_proj",
|
| 31 |
"in_proj",
|
|
|
|
|
|
|
|
|
|
| 32 |
"gate_proj",
|
| 33 |
-
"
|
|
|
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
+
"q_proj",
|
|
|
|
|
|
|
| 24 |
"v_proj",
|
|
|
|
| 25 |
"dt_proj",
|
| 26 |
+
"out_proj",
|
|
|
|
| 27 |
"in_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"x_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
"gate_proj",
|
| 32 |
+
"down_proj",
|
| 33 |
+
"up_proj",
|
| 34 |
"router"
|
| 35 |
],
|
| 36 |
"task_type": "CAUSAL_LM",
|
checkpoint-168/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 531611600
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f54d945abe2be9a1a659c9f6a5966e52d7ed377a90ad865b77df2ae56be278d1
|
| 3 |
size 531611600
|
checkpoint-168/global_step168/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 797643792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e4c09b43bea186099825610cb6037a20b5cdc7616c4b0c0cff35da8a99f88806
|
| 3 |
size 797643792
|
checkpoint-168/global_step168/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 797644432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6fde27562524ca3badefc0ac640cd2761d58191ac2180933190780a60570d85
|
| 3 |
size 797644432
|
checkpoint-168/global_step168/mp_rank_00_model_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1345660121
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f40c593123f55cdf67c443b279b6eac0236c47c74d4fd2671735a1e41a4f57e
|
| 3 |
size 1345660121
|
checkpoint-168/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ecd14f5486e42b398ac0fe7c317f329cc3f6385858e18676a22a6b18f29c226f
|
| 3 |
size 14512
|
checkpoint-168/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6b1c2e52201fc6231525ffe165db2aee0cbfcfdfcb4e2a61aac311b2649dde0
|
| 3 |
size 14512
|
checkpoint-168/trainer_state.json
CHANGED
|
@@ -10,7 +10,7 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
"loss": 0.6274,
|
| 16 |
"step": 1
|
|
@@ -18,1210 +18,1210 @@
|
|
| 18 |
{
|
| 19 |
"epoch": 0.01,
|
| 20 |
"eval_loss": 1.0297596454620361,
|
| 21 |
-
"eval_runtime":
|
| 22 |
-
"eval_samples_per_second": 1.
|
| 23 |
-
"eval_steps_per_second": 0.
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.01,
|
| 28 |
-
"grad_norm": 0.
|
| 29 |
"learning_rate": 4e-05,
|
| 30 |
"loss": 0.6362,
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.02,
|
| 35 |
-
"grad_norm": 0.
|
| 36 |
"learning_rate": 6e-05,
|
| 37 |
"loss": 0.6299,
|
| 38 |
"step": 3
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.02,
|
| 42 |
-
"grad_norm": 0.
|
| 43 |
"learning_rate": 8e-05,
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 4
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.03,
|
| 49 |
-
"grad_norm": 0.
|
| 50 |
"learning_rate": 0.0001,
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 5
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.04,
|
| 56 |
-
"grad_norm": 0.
|
| 57 |
"learning_rate": 0.00012,
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 6
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.04,
|
| 63 |
-
"grad_norm": 0.
|
| 64 |
"learning_rate": 0.00014,
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.05,
|
| 70 |
-
"grad_norm": 0.
|
| 71 |
"learning_rate": 0.00016,
|
| 72 |
-
"loss": 0.
|
| 73 |
"step": 8
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.05,
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
"learning_rate": 0.00018,
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 9
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.06,
|
| 84 |
-
"grad_norm": 0.
|
| 85 |
"learning_rate": 0.0002,
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 10
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.07,
|
| 91 |
-
"grad_norm": 0.
|
| 92 |
"learning_rate": 0.00019998023297700658,
|
| 93 |
-
"loss": 0.
|
| 94 |
"step": 11
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.07,
|
| 98 |
-
"grad_norm": 0.
|
| 99 |
"learning_rate": 0.00019992093972273018,
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 12
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.08,
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
"learning_rate": 0.00019982214367819328,
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 13
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.08,
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
"learning_rate": 0.0001996838839014696,
|
| 114 |
"loss": 0.5495,
|
| 115 |
"step": 14
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.09,
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
"learning_rate": 0.00019950621505224273,
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 15
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.1,
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
"learning_rate": 0.00019928920737019733,
|
| 128 |
"loss": 0.5083,
|
| 129 |
"step": 16
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.1,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
"learning_rate": 0.0001990329466472502,
|
| 135 |
-
"loss": 0.
|
| 136 |
"step": 17
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.11,
|
| 140 |
-
"grad_norm": 0.
|
| 141 |
"learning_rate": 0.00019873753419363336,
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 18
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.11,
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
"learning_rate": 0.00019840308679784207,
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 19
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.12,
|
| 154 |
-
"grad_norm": 0.
|
| 155 |
"learning_rate": 0.00019802973668046363,
|
| 156 |
-
"loss": 0.
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.12,
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
"learning_rate": 0.0001976176314419051,
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 21
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.13,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
"learning_rate": 0.000197166934004041,
|
| 170 |
"loss": 0.4819,
|
| 171 |
"step": 22
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.14,
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
"learning_rate": 0.00019667782254580374,
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 23
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.14,
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
"learning_rate": 0.00019615049043274205,
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 24
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15,
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
"learning_rate": 0.00019558514614057609,
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 25
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.15,
|
| 196 |
-
"grad_norm": 0.
|
| 197 |
"learning_rate": 0.00019498201317277828,
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 26
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.16,
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
"learning_rate": 0.00019434132997221345,
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 27
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.17,
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
"learning_rate": 0.0001936633498268728,
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 28
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.17,
|
| 217 |
-
"grad_norm": 0.
|
| 218 |
"learning_rate": 0.0001929483407697387,
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 29
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.18,
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
"learning_rate": 0.00019219658547282067,
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 30
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.18,
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
"learning_rate": 0.00019140838113540346,
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 31
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.19,
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
"learning_rate": 0.00019058403936655233,
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 32
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.2,
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
"learning_rate": 0.00018972388606192125,
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 33
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.2,
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
"learning_rate": 0.0001888282612749132,
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 34
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.21,
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
"learning_rate": 0.00018789751908224338,
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 35
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.21,
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
"learning_rate": 0.00018693202744395827,
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 36
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.22,
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
"learning_rate": 0.00018593216805796612,
|
| 275 |
"loss": 0.5396,
|
| 276 |
"step": 37
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.23,
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
"learning_rate": 0.00018489833620913642,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 38
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.23,
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
"learning_rate": 0.00018383094061302766,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 39
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.24,
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
"learning_rate": 0.00018273040325430574,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.24,
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
"learning_rate": 0.00018159715921991612,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 41
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.25,
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
"learning_rate": 0.00018043165652707649,
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 42
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.25,
|
| 315 |
-
"eval_loss": 0.
|
| 316 |
-
"eval_runtime":
|
| 317 |
-
"eval_samples_per_second": 1.
|
| 318 |
-
"eval_steps_per_second": 0.
|
| 319 |
"step": 42
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 0.26,
|
| 323 |
-
"grad_norm": 0.
|
| 324 |
"learning_rate": 0.00017923435594615744,
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 43
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 0.26,
|
| 330 |
-
"grad_norm": 0.
|
| 331 |
"learning_rate": 0.00017800573081852122,
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 44
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 0.27,
|
| 337 |
-
"grad_norm": 0.
|
| 338 |
"learning_rate": 0.0001767462668693908,
|
| 339 |
-
"loss": 0.
|
| 340 |
"step": 45
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 0.27,
|
| 344 |
-
"grad_norm": 0.
|
| 345 |
"learning_rate": 0.00017545646201582303,
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 46
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 0.28,
|
| 351 |
-
"grad_norm": 0.
|
| 352 |
"learning_rate": 0.00017413682616986185,
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 47
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.29,
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
"learning_rate": 0.00017278788103694943,
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 48
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.29,
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
"learning_rate": 0.000171410159909675,
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 49
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.3,
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
"learning_rate": 0.00017000420745694254,
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 50
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 0.3,
|
| 379 |
-
"grad_norm": 0.
|
| 380 |
"learning_rate": 0.00016857057950864132,
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 51
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 0.31,
|
| 386 |
-
"grad_norm": 0.
|
| 387 |
"learning_rate": 0.0001671098428359037,
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 52
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 0.32,
|
| 393 |
-
"grad_norm": 0.
|
| 394 |
"learning_rate": 0.00016562257492703757,
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 53
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.32,
|
| 400 |
-
"grad_norm": 0.
|
| 401 |
"learning_rate": 0.000164109363759222,
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 54
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.33,
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 0.000162570807566056,
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 55
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.33,
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
"learning_rate": 0.00016100751460105243,
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 56
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.34,
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
"learning_rate": 0.00015942010289717105,
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 57
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 0.34,
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
"learning_rate": 0.00015780920002248484,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 58
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.35,
|
| 435 |
-
"grad_norm": 0.
|
| 436 |
"learning_rate": 0.0001561754428320771,
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 59
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 0.36,
|
| 442 |
-
"grad_norm": 0.
|
| 443 |
"learning_rate": 0.00015451947721626676,
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 60
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 0.36,
|
| 449 |
-
"grad_norm": 0.
|
| 450 |
"learning_rate": 0.00015284195784526195,
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 61
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 0.37,
|
| 456 |
-
"grad_norm": 0.
|
| 457 |
"learning_rate": 0.00015114354791034225,
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 62
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.37,
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
"learning_rate": 0.0001494249188616723,
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 63
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 0.38,
|
| 470 |
-
"grad_norm": 0.
|
| 471 |
"learning_rate": 0.00014768675014285062,
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 64
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.39,
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
"learning_rate": 0.00014592972892229778,
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 65
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 0.39,
|
| 484 |
-
"grad_norm": 0.
|
| 485 |
"learning_rate": 0.0001441545498215912,
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 66
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 0.4,
|
| 491 |
-
"grad_norm": 0.
|
| 492 |
"learning_rate": 0.00014236191464085286,
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 67
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 0.4,
|
| 498 |
-
"grad_norm": 0.
|
| 499 |
"learning_rate": 0.00014055253208129938,
|
| 500 |
-
"loss": 0.
|
| 501 |
"step": 68
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 0.41,
|
| 505 |
-
"grad_norm": 0.
|
| 506 |
"learning_rate": 0.00013872711746506413,
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 69
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 0.42,
|
| 512 |
-
"grad_norm": 0.
|
| 513 |
"learning_rate": 0.00013688639245240078,
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 70
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 0.42,
|
| 519 |
-
"grad_norm": 0.
|
| 520 |
"learning_rate": 0.00013503108475638244,
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 71
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 0.43,
|
| 526 |
-
"grad_norm": 0.
|
| 527 |
"learning_rate": 0.0001331619278552068,
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 72
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.43,
|
| 533 |
-
"grad_norm": 0.
|
| 534 |
"learning_rate": 0.00013127966070222274,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 73
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 0.44,
|
| 540 |
-
"grad_norm": 0.
|
| 541 |
"learning_rate": 0.00012938502743379212,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 74
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 0.45,
|
| 547 |
-
"grad_norm": 0.
|
| 548 |
"learning_rate": 0.00012747877707510252,
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 75
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 0.45,
|
| 554 |
-
"grad_norm": 0.
|
| 555 |
"learning_rate": 0.0001255616632440475,
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 76
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 0.46,
|
| 561 |
-
"grad_norm": 0.
|
| 562 |
"learning_rate": 0.0001236344438532905,
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 77
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.46,
|
| 568 |
-
"grad_norm": 0.
|
| 569 |
"learning_rate": 0.0001216978808106318,
|
| 570 |
"loss": 0.5091,
|
| 571 |
"step": 78
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 0.47,
|
| 575 |
-
"grad_norm": 0.
|
| 576 |
"learning_rate": 0.00011975273971779528,
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 79
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.48,
|
| 582 |
-
"grad_norm": 0.
|
| 583 |
"learning_rate": 0.00011779978956775506,
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 80
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.48,
|
| 589 |
-
"grad_norm": 0.
|
| 590 |
"learning_rate": 0.0001158398024407215,
|
| 591 |
-
"loss": 0.
|
| 592 |
"step": 81
|
| 593 |
},
|
| 594 |
{
|
| 595 |
"epoch": 0.49,
|
| 596 |
-
"grad_norm": 0.
|
| 597 |
"learning_rate": 0.00011387355319890685,
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 82
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 0.49,
|
| 603 |
-
"grad_norm": 0.
|
| 604 |
"learning_rate": 0.00011190181918019049,
|
| 605 |
-
"loss": 0.
|
| 606 |
"step": 83
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 0.5,
|
| 610 |
-
"grad_norm": 0.
|
| 611 |
"learning_rate": 0.00010992537989080618,
|
| 612 |
-
"loss": 0.
|
| 613 |
"step": 84
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 0.5,
|
| 617 |
-
"eval_loss": 0.
|
| 618 |
-
"eval_runtime":
|
| 619 |
-
"eval_samples_per_second": 1.
|
| 620 |
-
"eval_steps_per_second": 0.
|
| 621 |
"step": 84
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 0.51,
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
"learning_rate": 0.00010794501669717145,
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 85
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 0.51,
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
"learning_rate": 0.00010596151251698199,
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 86
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 0.52,
|
| 639 |
-
"grad_norm": 0.
|
| 640 |
"learning_rate": 0.0001039756515096926,
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 87
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 0.52,
|
| 646 |
-
"grad_norm": 0.
|
| 647 |
"learning_rate": 0.00010198821876650701,
|
| 648 |
-
"loss": 0.
|
| 649 |
"step": 88
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.53,
|
| 653 |
-
"grad_norm": 0.
|
| 654 |
"learning_rate": 0.0001,
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 89
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 0.53,
|
| 660 |
-
"grad_norm": 0.
|
| 661 |
"learning_rate": 9.801178123349298e-05,
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 90
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 0.54,
|
| 667 |
-
"grad_norm": 0.
|
| 668 |
"learning_rate": 9.602434849030745e-05,
|
| 669 |
-
"loss": 0.
|
| 670 |
"step": 91
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 0.55,
|
| 674 |
-
"grad_norm": 0.
|
| 675 |
"learning_rate": 9.403848748301802e-05,
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 92
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 0.55,
|
| 681 |
-
"grad_norm": 0.
|
| 682 |
"learning_rate": 9.205498330282856e-05,
|
| 683 |
-
"loss": 0.
|
| 684 |
"step": 93
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.56,
|
| 688 |
-
"grad_norm": 0.
|
| 689 |
"learning_rate": 9.007462010919386e-05,
|
| 690 |
-
"loss": 0.
|
| 691 |
"step": 94
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 0.56,
|
| 695 |
-
"grad_norm": 0.
|
| 696 |
"learning_rate": 8.809818081980953e-05,
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 95
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 0.57,
|
| 702 |
-
"grad_norm": 0.
|
| 703 |
"learning_rate": 8.612644680109319e-05,
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 96
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.58,
|
| 709 |
-
"grad_norm": 0.
|
| 710 |
"learning_rate": 8.416019755927851e-05,
|
| 711 |
-
"loss": 0.
|
| 712 |
"step": 97
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 0.58,
|
| 716 |
-
"grad_norm": 0.
|
| 717 |
"learning_rate": 8.2200210432245e-05,
|
| 718 |
-
"loss": 0.
|
| 719 |
"step": 98
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 0.59,
|
| 723 |
-
"grad_norm": 0.
|
| 724 |
"learning_rate": 8.024726028220474e-05,
|
| 725 |
-
"loss": 0.
|
| 726 |
"step": 99
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 0.59,
|
| 730 |
-
"grad_norm": 0.
|
| 731 |
"learning_rate": 7.83021191893682e-05,
|
| 732 |
-
"loss": 0.
|
| 733 |
"step": 100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 0.6,
|
| 737 |
-
"grad_norm": 0.
|
| 738 |
"learning_rate": 7.636555614670953e-05,
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 101
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.61,
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
"learning_rate": 7.443833675595255e-05,
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 102
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.61,
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
"learning_rate": 7.252122292489747e-05,
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 103
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.62,
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
"learning_rate": 7.061497256620793e-05,
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 104
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.62,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 6.87203392977773e-05,
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 105
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.63,
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
"learning_rate": 6.683807214479323e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 106
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.64,
|
| 779 |
-
"grad_norm": 0.
|
| 780 |
"learning_rate": 6.496891524361757e-05,
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 107
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 0.64,
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
"learning_rate": 6.311360754759923e-05,
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 108
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.65,
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
"learning_rate": 6.127288253493591e-05,
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 109
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 0.65,
|
| 800 |
-
"grad_norm": 0.
|
| 801 |
"learning_rate": 5.9447467918700614e-05,
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 110
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 0.66,
|
| 807 |
-
"grad_norm": 0.
|
| 808 |
"learning_rate": 5.763808535914723e-05,
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 111
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 0.67,
|
| 814 |
-
"grad_norm": 0.
|
| 815 |
"learning_rate": 5.584545017840885e-05,
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 112
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.67,
|
| 821 |
-
"grad_norm": 0.
|
| 822 |
"learning_rate": 5.407027107770219e-05,
|
| 823 |
-
"loss": 0.
|
| 824 |
"step": 113
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 0.68,
|
| 828 |
-
"grad_norm": 0.
|
| 829 |
"learning_rate": 5.2313249857149414e-05,
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 114
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 0.68,
|
| 835 |
-
"grad_norm": 0.
|
| 836 |
"learning_rate": 5.0575081138327715e-05,
|
| 837 |
-
"loss": 0.
|
| 838 |
"step": 115
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 0.69,
|
| 842 |
-
"grad_norm": 0.
|
| 843 |
"learning_rate": 4.885645208965779e-05,
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 116
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 0.7,
|
| 849 |
-
"grad_norm": 0.
|
| 850 |
"learning_rate": 4.715804215473809e-05,
|
| 851 |
-
"loss": 0.
|
| 852 |
"step": 117
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 0.7,
|
| 856 |
-
"grad_norm": 0.
|
| 857 |
"learning_rate": 4.548052278373327e-05,
|
| 858 |
-
"loss": 0.
|
| 859 |
"step": 118
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 0.71,
|
| 863 |
-
"grad_norm": 0.
|
| 864 |
"learning_rate": 4.382455716792291e-05,
|
| 865 |
-
"loss": 0.
|
| 866 |
"step": 119
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 0.71,
|
| 870 |
-
"grad_norm": 0.
|
| 871 |
"learning_rate": 4.219079997751515e-05,
|
| 872 |
"loss": 0.4954,
|
| 873 |
"step": 120
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.72,
|
| 877 |
-
"grad_norm": 0.
|
| 878 |
"learning_rate": 4.0579897102828966e-05,
|
| 879 |
-
"loss": 0.
|
| 880 |
"step": 121
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 0.73,
|
| 884 |
-
"grad_norm": 0.
|
| 885 |
"learning_rate": 3.899248539894757e-05,
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 122
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 0.73,
|
| 891 |
-
"grad_norm": 0.
|
| 892 |
"learning_rate": 3.7429192433944014e-05,
|
| 893 |
-
"loss": 0.
|
| 894 |
"step": 123
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.74,
|
| 898 |
-
"grad_norm": 0.
|
| 899 |
"learning_rate": 3.589063624077802e-05,
|
| 900 |
-
"loss": 0.
|
| 901 |
"step": 124
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.74,
|
| 905 |
-
"grad_norm": 0.
|
| 906 |
"learning_rate": 3.4377425072962465e-05,
|
| 907 |
-
"loss": 0.
|
| 908 |
"step": 125
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 0.75,
|
| 912 |
-
"grad_norm": 0.
|
| 913 |
"learning_rate": 3.289015716409631e-05,
|
| 914 |
-
"loss": 0.
|
| 915 |
"step": 126
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 0.75,
|
| 919 |
-
"eval_loss": 0.
|
| 920 |
-
"eval_runtime": 155.
|
| 921 |
-
"eval_samples_per_second": 1.
|
| 922 |
-
"eval_steps_per_second": 0.
|
| 923 |
"step": 126
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 0.75,
|
| 927 |
-
"grad_norm": 0.
|
| 928 |
"learning_rate": 3.14294204913587e-05,
|
| 929 |
-
"loss": 0.
|
| 930 |
"step": 127
|
| 931 |
},
|
| 932 |
{
|
| 933 |
"epoch": 0.76,
|
| 934 |
-
"grad_norm": 0.
|
| 935 |
"learning_rate": 2.9995792543057478e-05,
|
| 936 |
-
"loss": 0.
|
| 937 |
"step": 128
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 0.77,
|
| 941 |
-
"grad_norm": 0.
|
| 942 |
"learning_rate": 2.8589840090325027e-05,
|
| 943 |
-
"loss": 0.
|
| 944 |
"step": 129
|
| 945 |
},
|
| 946 |
{
|
| 947 |
"epoch": 0.77,
|
| 948 |
-
"grad_norm": 0.
|
| 949 |
"learning_rate": 2.7212118963050592e-05,
|
| 950 |
-
"loss": 0.
|
| 951 |
"step": 130
|
| 952 |
},
|
| 953 |
{
|
| 954 |
"epoch": 0.78,
|
| 955 |
-
"grad_norm": 0.
|
| 956 |
"learning_rate": 2.586317383013821e-05,
|
| 957 |
-
"loss": 0.
|
| 958 |
"step": 131
|
| 959 |
},
|
| 960 |
{
|
| 961 |
"epoch": 0.78,
|
| 962 |
-
"grad_norm": 0.
|
| 963 |
"learning_rate": 2.4543537984176978e-05,
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 132
|
| 966 |
},
|
| 967 |
{
|
| 968 |
"epoch": 0.79,
|
| 969 |
-
"grad_norm": 0.
|
| 970 |
"learning_rate": 2.325373313060919e-05,
|
| 971 |
-
"loss": 0.
|
| 972 |
"step": 133
|
| 973 |
},
|
| 974 |
{
|
| 975 |
"epoch": 0.8,
|
| 976 |
-
"grad_norm": 0.
|
| 977 |
"learning_rate": 2.19942691814788e-05,
|
| 978 |
-
"loss": 0.
|
| 979 |
"step": 134
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 0.8,
|
| 983 |
-
"grad_norm": 0.
|
| 984 |
"learning_rate": 2.076564405384258e-05,
|
| 985 |
-
"loss": 0.
|
| 986 |
"step": 135
|
| 987 |
},
|
| 988 |
{
|
| 989 |
"epoch": 0.81,
|
| 990 |
-
"grad_norm": 0.
|
| 991 |
"learning_rate": 1.9568343472923524e-05,
|
| 992 |
-
"loss": 0.
|
| 993 |
"step": 136
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 0.81,
|
| 997 |
-
"grad_norm": 0.
|
| 998 |
"learning_rate": 1.840284078008393e-05,
|
| 999 |
-
"loss": 0.
|
| 1000 |
"step": 137
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
"epoch": 0.82,
|
| 1004 |
-
"grad_norm": 0.
|
| 1005 |
"learning_rate": 1.7269596745694295e-05,
|
| 1006 |
-
"loss": 0.
|
| 1007 |
"step": 138
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 0.83,
|
| 1011 |
-
"grad_norm": 0.
|
| 1012 |
"learning_rate": 1.616905938697234e-05,
|
| 1013 |
-
"loss": 0.
|
| 1014 |
"step": 139
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 0.83,
|
| 1018 |
-
"grad_norm": 0.
|
| 1019 |
"learning_rate": 1.5101663790863596e-05,
|
| 1020 |
-
"loss": 0.
|
| 1021 |
"step": 140
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 0.84,
|
| 1025 |
-
"grad_norm": 0.
|
| 1026 |
"learning_rate": 1.4067831942033904e-05,
|
| 1027 |
-
"loss": 0.
|
| 1028 |
"step": 141
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 0.84,
|
| 1032 |
-
"grad_norm": 0.
|
| 1033 |
"learning_rate": 1.3067972556041752e-05,
|
| 1034 |
-
"loss": 0.
|
| 1035 |
"step": 142
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
"epoch": 0.85,
|
| 1039 |
-
"grad_norm": 0.
|
| 1040 |
"learning_rate": 1.210248091775663e-05,
|
| 1041 |
-
"loss": 0.
|
| 1042 |
"step": 143
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 0.86,
|
| 1046 |
-
"grad_norm": 0.
|
| 1047 |
"learning_rate": 1.1171738725086833e-05,
|
| 1048 |
-
"loss": 0.
|
| 1049 |
"step": 144
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 0.86,
|
| 1053 |
-
"grad_norm": 0.
|
| 1054 |
"learning_rate": 1.0276113938078769e-05,
|
| 1055 |
-
"loss": 0.
|
| 1056 |
"step": 145
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 0.87,
|
| 1060 |
-
"grad_norm": 0.
|
| 1061 |
"learning_rate": 9.415960633447674e-06,
|
| 1062 |
"loss": 0.4452,
|
| 1063 |
"step": 146
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 0.87,
|
| 1067 |
-
"grad_norm": 0.
|
| 1068 |
"learning_rate": 8.59161886459654e-06,
|
| 1069 |
-
"loss": 0.
|
| 1070 |
"step": 147
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
"epoch": 0.88,
|
| 1074 |
-
"grad_norm": 0.
|
| 1075 |
"learning_rate": 7.803414527179343e-06,
|
| 1076 |
-
"loss": 0.
|
| 1077 |
"step": 148
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
"epoch": 0.89,
|
| 1081 |
-
"grad_norm": 0.
|
| 1082 |
"learning_rate": 7.051659230261298e-06,
|
| 1083 |
-
"loss": 0.
|
| 1084 |
"step": 149
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 0.89,
|
| 1088 |
-
"grad_norm": 0.
|
| 1089 |
"learning_rate": 6.336650173127223e-06,
|
| 1090 |
-
"loss": 0.
|
| 1091 |
"step": 150
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 0.9,
|
| 1095 |
-
"grad_norm": 0.
|
| 1096 |
"learning_rate": 5.658670027786561e-06,
|
| 1097 |
-
"loss": 0.
|
| 1098 |
"step": 151
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
"epoch": 0.9,
|
| 1102 |
-
"grad_norm": 0.
|
| 1103 |
"learning_rate": 5.017986827221733e-06,
|
| 1104 |
-
"loss": 0.
|
| 1105 |
"step": 152
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 0.91,
|
| 1109 |
-
"grad_norm": 0.
|
| 1110 |
"learning_rate": 4.4148538594239174e-06,
|
| 1111 |
-
"loss": 0.
|
| 1112 |
"step": 153
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 0.92,
|
| 1116 |
-
"grad_norm": 0.
|
| 1117 |
"learning_rate": 3.849509567257959e-06,
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 154
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 0.92,
|
| 1123 |
-
"grad_norm": 0.
|
| 1124 |
"learning_rate": 3.3221774541962845e-06,
|
| 1125 |
-
"loss": 0.
|
| 1126 |
"step": 155
|
| 1127 |
},
|
| 1128 |
{
|
| 1129 |
"epoch": 0.93,
|
| 1130 |
-
"grad_norm": 0.
|
| 1131 |
"learning_rate": 2.8330659959589946e-06,
|
| 1132 |
-
"loss": 0.
|
| 1133 |
"step": 156
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
"epoch": 0.93,
|
| 1137 |
-
"grad_norm": 0.
|
| 1138 |
"learning_rate": 2.3823685580949273e-06,
|
| 1139 |
-
"loss": 0.
|
| 1140 |
"step": 157
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
"epoch": 0.94,
|
| 1144 |
-
"grad_norm": 0.
|
| 1145 |
"learning_rate": 1.9702633195363917e-06,
|
| 1146 |
-
"loss": 0.
|
| 1147 |
"step": 158
|
| 1148 |
},
|
| 1149 |
{
|
| 1150 |
"epoch": 0.95,
|
| 1151 |
-
"grad_norm": 0.
|
| 1152 |
"learning_rate": 1.5969132021579347e-06,
|
| 1153 |
-
"loss": 0.
|
| 1154 |
"step": 159
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
"epoch": 0.95,
|
| 1158 |
-
"grad_norm": 0.
|
| 1159 |
"learning_rate": 1.2624658063666639e-06,
|
| 1160 |
-
"loss": 0.
|
| 1161 |
"step": 160
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
"epoch": 0.96,
|
| 1165 |
-
"grad_norm": 0.
|
| 1166 |
"learning_rate": 9.670533527498137e-07,
|
| 1167 |
-
"loss": 0.
|
| 1168 |
"step": 161
|
| 1169 |
},
|
| 1170 |
{
|
| 1171 |
"epoch": 0.96,
|
| 1172 |
-
"grad_norm": 0.
|
| 1173 |
"learning_rate": 7.10792629802659e-07,
|
| 1174 |
-
"loss": 0.
|
| 1175 |
"step": 162
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
"epoch": 0.97,
|
| 1179 |
-
"grad_norm": 0.
|
| 1180 |
"learning_rate": 4.937849477572587e-07,
|
| 1181 |
-
"loss": 0.
|
| 1182 |
"step": 163
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
"epoch": 0.97,
|
| 1186 |
-
"grad_norm": 0.
|
| 1187 |
"learning_rate": 3.161160985304168e-07,
|
| 1188 |
-
"loss": 0.
|
| 1189 |
"step": 164
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
"epoch": 0.98,
|
| 1193 |
-
"grad_norm": 0.
|
| 1194 |
"learning_rate": 1.7785632180670198e-07,
|
| 1195 |
"loss": 0.436,
|
| 1196 |
"step": 165
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
"epoch": 0.99,
|
| 1200 |
-
"grad_norm": 0.
|
| 1201 |
"learning_rate": 7.906027726981568e-08,
|
| 1202 |
-
"loss": 0.
|
| 1203 |
"step": 166
|
| 1204 |
},
|
| 1205 |
{
|
| 1206 |
"epoch": 0.99,
|
| 1207 |
-
"grad_norm": 0.
|
| 1208 |
"learning_rate": 1.976702299344435e-08,
|
| 1209 |
-
"loss": 0.
|
| 1210 |
"step": 167
|
| 1211 |
},
|
| 1212 |
{
|
| 1213 |
"epoch": 1.0,
|
| 1214 |
-
"grad_norm": 0.
|
| 1215 |
"learning_rate": 0.0,
|
| 1216 |
-
"loss": 0.
|
| 1217 |
"step": 168
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 1.0,
|
| 1221 |
-
"eval_loss": 0.
|
| 1222 |
-
"eval_runtime": 154.
|
| 1223 |
-
"eval_samples_per_second": 1.
|
| 1224 |
-
"eval_steps_per_second": 0.
|
| 1225 |
"step": 168
|
| 1226 |
}
|
| 1227 |
],
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.29289755909939047,
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
"loss": 0.6274,
|
| 16 |
"step": 1
|
|
|
|
| 18 |
{
|
| 19 |
"epoch": 0.01,
|
| 20 |
"eval_loss": 1.0297596454620361,
|
| 21 |
+
"eval_runtime": 153.0715,
|
| 22 |
+
"eval_samples_per_second": 1.627,
|
| 23 |
+
"eval_steps_per_second": 0.817,
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.01,
|
| 28 |
+
"grad_norm": 0.27956410941469395,
|
| 29 |
"learning_rate": 4e-05,
|
| 30 |
"loss": 0.6362,
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.02,
|
| 35 |
+
"grad_norm": 0.3009590515152092,
|
| 36 |
"learning_rate": 6e-05,
|
| 37 |
"loss": 0.6299,
|
| 38 |
"step": 3
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.02,
|
| 42 |
+
"grad_norm": 0.34354546270515235,
|
| 43 |
"learning_rate": 8e-05,
|
| 44 |
+
"loss": 0.6395,
|
| 45 |
"step": 4
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.03,
|
| 49 |
+
"grad_norm": 0.2470961998205002,
|
| 50 |
"learning_rate": 0.0001,
|
| 51 |
+
"loss": 0.6068,
|
| 52 |
"step": 5
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.04,
|
| 56 |
+
"grad_norm": 0.2071993792912813,
|
| 57 |
"learning_rate": 0.00012,
|
| 58 |
+
"loss": 0.5993,
|
| 59 |
"step": 6
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.04,
|
| 63 |
+
"grad_norm": 0.20237019487582247,
|
| 64 |
"learning_rate": 0.00014,
|
| 65 |
+
"loss": 0.6293,
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.05,
|
| 70 |
+
"grad_norm": 0.13810925455451734,
|
| 71 |
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 0.5101,
|
| 73 |
"step": 8
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.05,
|
| 77 |
+
"grad_norm": 0.22762469698117493,
|
| 78 |
"learning_rate": 0.00018,
|
| 79 |
+
"loss": 0.5527,
|
| 80 |
"step": 9
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.06,
|
| 84 |
+
"grad_norm": 0.15718586910417978,
|
| 85 |
"learning_rate": 0.0002,
|
| 86 |
+
"loss": 0.5437,
|
| 87 |
"step": 10
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.07,
|
| 91 |
+
"grad_norm": 0.13825435646308676,
|
| 92 |
"learning_rate": 0.00019998023297700658,
|
| 93 |
+
"loss": 0.5858,
|
| 94 |
"step": 11
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 0.07,
|
| 98 |
+
"grad_norm": 0.15072585803075744,
|
| 99 |
"learning_rate": 0.00019992093972273018,
|
| 100 |
+
"loss": 0.554,
|
| 101 |
"step": 12
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 0.08,
|
| 105 |
+
"grad_norm": 0.13616736806741778,
|
| 106 |
"learning_rate": 0.00019982214367819328,
|
| 107 |
+
"loss": 0.6199,
|
| 108 |
"step": 13
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 0.08,
|
| 112 |
+
"grad_norm": 0.15841398198140327,
|
| 113 |
"learning_rate": 0.0001996838839014696,
|
| 114 |
"loss": 0.5495,
|
| 115 |
"step": 14
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.09,
|
| 119 |
+
"grad_norm": 0.16548492090565414,
|
| 120 |
"learning_rate": 0.00019950621505224273,
|
| 121 |
+
"loss": 0.5043,
|
| 122 |
"step": 15
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.1,
|
| 126 |
+
"grad_norm": 0.16965923620151113,
|
| 127 |
"learning_rate": 0.00019928920737019733,
|
| 128 |
"loss": 0.5083,
|
| 129 |
"step": 16
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.1,
|
| 133 |
+
"grad_norm": 0.1853247768877015,
|
| 134 |
"learning_rate": 0.0001990329466472502,
|
| 135 |
+
"loss": 0.6318,
|
| 136 |
"step": 17
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.11,
|
| 140 |
+
"grad_norm": 0.1824469971767483,
|
| 141 |
"learning_rate": 0.00019873753419363336,
|
| 142 |
+
"loss": 0.4809,
|
| 143 |
"step": 18
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.11,
|
| 147 |
+
"grad_norm": 0.13288488939703025,
|
| 148 |
"learning_rate": 0.00019840308679784207,
|
| 149 |
+
"loss": 0.4974,
|
| 150 |
"step": 19
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 0.12,
|
| 154 |
+
"grad_norm": 0.18507602579352214,
|
| 155 |
"learning_rate": 0.00019802973668046363,
|
| 156 |
+
"loss": 0.5288,
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.12,
|
| 161 |
+
"grad_norm": 0.13591012925799634,
|
| 162 |
"learning_rate": 0.0001976176314419051,
|
| 163 |
+
"loss": 0.5291,
|
| 164 |
"step": 21
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.13,
|
| 168 |
+
"grad_norm": 0.13704233024715257,
|
| 169 |
"learning_rate": 0.000197166934004041,
|
| 170 |
"loss": 0.4819,
|
| 171 |
"step": 22
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.14,
|
| 175 |
+
"grad_norm": 0.15362008062157553,
|
| 176 |
"learning_rate": 0.00019667782254580374,
|
| 177 |
+
"loss": 0.5408,
|
| 178 |
"step": 23
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 0.14,
|
| 182 |
+
"grad_norm": 0.11171032733778614,
|
| 183 |
"learning_rate": 0.00019615049043274205,
|
| 184 |
+
"loss": 0.5101,
|
| 185 |
"step": 24
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.15,
|
| 189 |
+
"grad_norm": 0.11449051252849428,
|
| 190 |
"learning_rate": 0.00019558514614057609,
|
| 191 |
+
"loss": 0.5209,
|
| 192 |
"step": 25
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 0.15,
|
| 196 |
+
"grad_norm": 0.11611407915744347,
|
| 197 |
"learning_rate": 0.00019498201317277828,
|
| 198 |
+
"loss": 0.5005,
|
| 199 |
"step": 26
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 0.16,
|
| 203 |
+
"grad_norm": 0.12735633641627706,
|
| 204 |
"learning_rate": 0.00019434132997221345,
|
| 205 |
+
"loss": 0.4741,
|
| 206 |
"step": 27
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 0.17,
|
| 210 |
+
"grad_norm": 0.11871518327376328,
|
| 211 |
"learning_rate": 0.0001936633498268728,
|
| 212 |
+
"loss": 0.5213,
|
| 213 |
"step": 28
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 0.17,
|
| 217 |
+
"grad_norm": 0.11403376465541806,
|
| 218 |
"learning_rate": 0.0001929483407697387,
|
| 219 |
+
"loss": 0.4842,
|
| 220 |
"step": 29
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.18,
|
| 224 |
+
"grad_norm": 0.11675155934307391,
|
| 225 |
"learning_rate": 0.00019219658547282067,
|
| 226 |
+
"loss": 0.4825,
|
| 227 |
"step": 30
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.18,
|
| 231 |
+
"grad_norm": 0.10789338581384152,
|
| 232 |
"learning_rate": 0.00019140838113540346,
|
| 233 |
+
"loss": 0.4866,
|
| 234 |
"step": 31
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.19,
|
| 238 |
+
"grad_norm": 0.14336837303756964,
|
| 239 |
"learning_rate": 0.00019058403936655233,
|
| 240 |
+
"loss": 0.5325,
|
| 241 |
"step": 32
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 0.2,
|
| 245 |
+
"grad_norm": 0.10401694793599091,
|
| 246 |
"learning_rate": 0.00018972388606192125,
|
| 247 |
+
"loss": 0.4292,
|
| 248 |
"step": 33
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 0.2,
|
| 252 |
+
"grad_norm": 0.10614245315138679,
|
| 253 |
"learning_rate": 0.0001888282612749132,
|
| 254 |
+
"loss": 0.4638,
|
| 255 |
"step": 34
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.21,
|
| 259 |
+
"grad_norm": 0.1250509143492961,
|
| 260 |
"learning_rate": 0.00018789751908224338,
|
| 261 |
+
"loss": 0.4963,
|
| 262 |
"step": 35
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 0.21,
|
| 266 |
+
"grad_norm": 0.12668831423294083,
|
| 267 |
"learning_rate": 0.00018693202744395827,
|
| 268 |
+
"loss": 0.5043,
|
| 269 |
"step": 36
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 0.22,
|
| 273 |
+
"grad_norm": 0.11594924793976216,
|
| 274 |
"learning_rate": 0.00018593216805796612,
|
| 275 |
"loss": 0.5396,
|
| 276 |
"step": 37
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 0.23,
|
| 280 |
+
"grad_norm": 0.12365894749489846,
|
| 281 |
"learning_rate": 0.00018489833620913642,
|
| 282 |
+
"loss": 0.4899,
|
| 283 |
"step": 38
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.23,
|
| 287 |
+
"grad_norm": 0.13164318970805183,
|
| 288 |
"learning_rate": 0.00018383094061302766,
|
| 289 |
+
"loss": 0.5065,
|
| 290 |
"step": 39
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.24,
|
| 294 |
+
"grad_norm": 0.10587870431925453,
|
| 295 |
"learning_rate": 0.00018273040325430574,
|
| 296 |
+
"loss": 0.4805,
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 0.24,
|
| 301 |
+
"grad_norm": 0.12134503625595237,
|
| 302 |
"learning_rate": 0.00018159715921991612,
|
| 303 |
+
"loss": 0.5103,
|
| 304 |
"step": 41
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 0.25,
|
| 308 |
+
"grad_norm": 0.10855490758268896,
|
| 309 |
"learning_rate": 0.00018043165652707649,
|
| 310 |
+
"loss": 0.44,
|
| 311 |
"step": 42
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.25,
|
| 315 |
+
"eval_loss": 0.9770342707633972,
|
| 316 |
+
"eval_runtime": 154.3933,
|
| 317 |
+
"eval_samples_per_second": 1.613,
|
| 318 |
+
"eval_steps_per_second": 0.81,
|
| 319 |
"step": 42
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 0.26,
|
| 323 |
+
"grad_norm": 0.11753509152196469,
|
| 324 |
"learning_rate": 0.00017923435594615744,
|
| 325 |
+
"loss": 0.4819,
|
| 326 |
"step": 43
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 0.26,
|
| 330 |
+
"grad_norm": 0.13774603727779988,
|
| 331 |
"learning_rate": 0.00017800573081852122,
|
| 332 |
+
"loss": 0.5451,
|
| 333 |
"step": 44
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 0.27,
|
| 337 |
+
"grad_norm": 0.11994636996852912,
|
| 338 |
"learning_rate": 0.0001767462668693908,
|
| 339 |
+
"loss": 0.5079,
|
| 340 |
"step": 45
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 0.27,
|
| 344 |
+
"grad_norm": 0.11803018063108017,
|
| 345 |
"learning_rate": 0.00017545646201582303,
|
| 346 |
+
"loss": 0.5183,
|
| 347 |
"step": 46
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 0.28,
|
| 351 |
+
"grad_norm": 0.12122026879022209,
|
| 352 |
"learning_rate": 0.00017413682616986185,
|
| 353 |
+
"loss": 0.4692,
|
| 354 |
"step": 47
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 0.29,
|
| 358 |
+
"grad_norm": 0.12840154129375927,
|
| 359 |
"learning_rate": 0.00017278788103694943,
|
| 360 |
+
"loss": 0.4538,
|
| 361 |
"step": 48
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 0.29,
|
| 365 |
+
"grad_norm": 0.10754191454075242,
|
| 366 |
"learning_rate": 0.000171410159909675,
|
| 367 |
+
"loss": 0.4745,
|
| 368 |
"step": 49
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.3,
|
| 372 |
+
"grad_norm": 0.10980392972154758,
|
| 373 |
"learning_rate": 0.00017000420745694254,
|
| 374 |
+
"loss": 0.5077,
|
| 375 |
"step": 50
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 0.3,
|
| 379 |
+
"grad_norm": 0.1018314102740997,
|
| 380 |
"learning_rate": 0.00016857057950864132,
|
| 381 |
+
"loss": 0.5077,
|
| 382 |
"step": 51
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 0.31,
|
| 386 |
+
"grad_norm": 0.10572512690181787,
|
| 387 |
"learning_rate": 0.0001671098428359037,
|
| 388 |
+
"loss": 0.4637,
|
| 389 |
"step": 52
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 0.32,
|
| 393 |
+
"grad_norm": 0.13106813432864392,
|
| 394 |
"learning_rate": 0.00016562257492703757,
|
| 395 |
+
"loss": 0.4718,
|
| 396 |
"step": 53
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 0.32,
|
| 400 |
+
"grad_norm": 0.11024484284605006,
|
| 401 |
"learning_rate": 0.000164109363759222,
|
| 402 |
+
"loss": 0.5115,
|
| 403 |
"step": 54
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 0.33,
|
| 407 |
+
"grad_norm": 0.12388815222110366,
|
| 408 |
"learning_rate": 0.000162570807566056,
|
| 409 |
+
"loss": 0.4756,
|
| 410 |
"step": 55
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 0.33,
|
| 414 |
+
"grad_norm": 0.12702188830349206,
|
| 415 |
"learning_rate": 0.00016100751460105243,
|
| 416 |
+
"loss": 0.4881,
|
| 417 |
"step": 56
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 0.34,
|
| 421 |
+
"grad_norm": 0.11086308659932648,
|
| 422 |
"learning_rate": 0.00015942010289717105,
|
| 423 |
+
"loss": 0.4701,
|
| 424 |
"step": 57
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 0.34,
|
| 428 |
+
"grad_norm": 0.11517890306226485,
|
| 429 |
"learning_rate": 0.00015780920002248484,
|
| 430 |
+
"loss": 0.4835,
|
| 431 |
"step": 58
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.35,
|
| 435 |
+
"grad_norm": 0.1253541013817757,
|
| 436 |
"learning_rate": 0.0001561754428320771,
|
| 437 |
+
"loss": 0.5147,
|
| 438 |
"step": 59
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 0.36,
|
| 442 |
+
"grad_norm": 0.11139862438389284,
|
| 443 |
"learning_rate": 0.00015451947721626676,
|
| 444 |
+
"loss": 0.4552,
|
| 445 |
"step": 60
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 0.36,
|
| 449 |
+
"grad_norm": 0.1152466912880715,
|
| 450 |
"learning_rate": 0.00015284195784526195,
|
| 451 |
+
"loss": 0.5053,
|
| 452 |
"step": 61
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 0.37,
|
| 456 |
+
"grad_norm": 0.11719477948785388,
|
| 457 |
"learning_rate": 0.00015114354791034225,
|
| 458 |
+
"loss": 0.5079,
|
| 459 |
"step": 62
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 0.37,
|
| 463 |
+
"grad_norm": 0.12730957754598396,
|
| 464 |
"learning_rate": 0.0001494249188616723,
|
| 465 |
+
"loss": 0.58,
|
| 466 |
"step": 63
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 0.38,
|
| 470 |
+
"grad_norm": 0.11907511614241814,
|
| 471 |
"learning_rate": 0.00014768675014285062,
|
| 472 |
+
"loss": 0.4579,
|
| 473 |
"step": 64
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 0.39,
|
| 477 |
+
"grad_norm": 0.13613222195950112,
|
| 478 |
"learning_rate": 0.00014592972892229778,
|
| 479 |
+
"loss": 0.4965,
|
| 480 |
"step": 65
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 0.39,
|
| 484 |
+
"grad_norm": 0.11610538032286319,
|
| 485 |
"learning_rate": 0.0001441545498215912,
|
| 486 |
+
"loss": 0.4467,
|
| 487 |
"step": 66
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 0.4,
|
| 491 |
+
"grad_norm": 0.1046724680527729,
|
| 492 |
"learning_rate": 0.00014236191464085286,
|
| 493 |
+
"loss": 0.4463,
|
| 494 |
"step": 67
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 0.4,
|
| 498 |
+
"grad_norm": 0.11631551012538931,
|
| 499 |
"learning_rate": 0.00014055253208129938,
|
| 500 |
+
"loss": 0.5304,
|
| 501 |
"step": 68
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 0.41,
|
| 505 |
+
"grad_norm": 0.11792788666231199,
|
| 506 |
"learning_rate": 0.00013872711746506413,
|
| 507 |
+
"loss": 0.453,
|
| 508 |
"step": 69
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 0.42,
|
| 512 |
+
"grad_norm": 0.11566019183159004,
|
| 513 |
"learning_rate": 0.00013688639245240078,
|
| 514 |
+
"loss": 0.5192,
|
| 515 |
"step": 70
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 0.42,
|
| 519 |
+
"grad_norm": 0.12967661214418025,
|
| 520 |
"learning_rate": 0.00013503108475638244,
|
| 521 |
+
"loss": 0.5625,
|
| 522 |
"step": 71
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 0.43,
|
| 526 |
+
"grad_norm": 0.10928605499181634,
|
| 527 |
"learning_rate": 0.0001331619278552068,
|
| 528 |
+
"loss": 0.4861,
|
| 529 |
"step": 72
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 0.43,
|
| 533 |
+
"grad_norm": 0.12485741141890881,
|
| 534 |
"learning_rate": 0.00013127966070222274,
|
| 535 |
+
"loss": 0.4782,
|
| 536 |
"step": 73
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 0.44,
|
| 540 |
+
"grad_norm": 0.1271574869759,
|
| 541 |
"learning_rate": 0.00012938502743379212,
|
| 542 |
+
"loss": 0.4819,
|
| 543 |
"step": 74
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 0.45,
|
| 547 |
+
"grad_norm": 0.14528941719728583,
|
| 548 |
"learning_rate": 0.00012747877707510252,
|
| 549 |
+
"loss": 0.5132,
|
| 550 |
"step": 75
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 0.45,
|
| 554 |
+
"grad_norm": 0.12760881412243183,
|
| 555 |
"learning_rate": 0.0001255616632440475,
|
| 556 |
+
"loss": 0.4817,
|
| 557 |
"step": 76
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 0.46,
|
| 561 |
+
"grad_norm": 0.1239773484446177,
|
| 562 |
"learning_rate": 0.0001236344438532905,
|
| 563 |
+
"loss": 0.5045,
|
| 564 |
"step": 77
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 0.46,
|
| 568 |
+
"grad_norm": 0.1307497276864945,
|
| 569 |
"learning_rate": 0.0001216978808106318,
|
| 570 |
"loss": 0.5091,
|
| 571 |
"step": 78
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 0.47,
|
| 575 |
+
"grad_norm": 0.1355749101256534,
|
| 576 |
"learning_rate": 0.00011975273971779528,
|
| 577 |
+
"loss": 0.5145,
|
| 578 |
"step": 79
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.48,
|
| 582 |
+
"grad_norm": 0.12394948316350356,
|
| 583 |
"learning_rate": 0.00011779978956775506,
|
| 584 |
+
"loss": 0.5057,
|
| 585 |
"step": 80
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 0.48,
|
| 589 |
+
"grad_norm": 0.13424789799433426,
|
| 590 |
"learning_rate": 0.0001158398024407215,
|
| 591 |
+
"loss": 0.5058,
|
| 592 |
"step": 81
|
| 593 |
},
|
| 594 |
{
|
| 595 |
"epoch": 0.49,
|
| 596 |
+
"grad_norm": 0.13054427599188898,
|
| 597 |
"learning_rate": 0.00011387355319890685,
|
| 598 |
+
"loss": 0.4683,
|
| 599 |
"step": 82
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 0.49,
|
| 603 |
+
"grad_norm": 0.11996704114604598,
|
| 604 |
"learning_rate": 0.00011190181918019049,
|
| 605 |
+
"loss": 0.4748,
|
| 606 |
"step": 83
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 0.5,
|
| 610 |
+
"grad_norm": 0.13071208464837492,
|
| 611 |
"learning_rate": 0.00010992537989080618,
|
| 612 |
+
"loss": 0.4406,
|
| 613 |
"step": 84
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"epoch": 0.5,
|
| 617 |
+
"eval_loss": 0.9652944803237915,
|
| 618 |
+
"eval_runtime": 153.5377,
|
| 619 |
+
"eval_samples_per_second": 1.622,
|
| 620 |
+
"eval_steps_per_second": 0.814,
|
| 621 |
"step": 84
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 0.51,
|
| 625 |
+
"grad_norm": 0.11393059638442392,
|
| 626 |
"learning_rate": 0.00010794501669717145,
|
| 627 |
+
"loss": 0.4877,
|
| 628 |
"step": 85
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 0.51,
|
| 632 |
+
"grad_norm": 0.11364636282385192,
|
| 633 |
"learning_rate": 0.00010596151251698199,
|
| 634 |
+
"loss": 0.4597,
|
| 635 |
"step": 86
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 0.52,
|
| 639 |
+
"grad_norm": 0.12548992883216656,
|
| 640 |
"learning_rate": 0.0001039756515096926,
|
| 641 |
+
"loss": 0.4685,
|
| 642 |
"step": 87
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 0.52,
|
| 646 |
+
"grad_norm": 0.12825123946992636,
|
| 647 |
"learning_rate": 0.00010198821876650701,
|
| 648 |
+
"loss": 0.4924,
|
| 649 |
"step": 88
|
| 650 |
},
|
| 651 |
{
|
| 652 |
"epoch": 0.53,
|
| 653 |
+
"grad_norm": 0.12264617890724591,
|
| 654 |
"learning_rate": 0.0001,
|
| 655 |
+
"loss": 0.4678,
|
| 656 |
"step": 89
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 0.53,
|
| 660 |
+
"grad_norm": 0.1238990144553216,
|
| 661 |
"learning_rate": 9.801178123349298e-05,
|
| 662 |
+
"loss": 0.4854,
|
| 663 |
"step": 90
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 0.54,
|
| 667 |
+
"grad_norm": 0.13533454171482565,
|
| 668 |
"learning_rate": 9.602434849030745e-05,
|
| 669 |
+
"loss": 0.4784,
|
| 670 |
"step": 91
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 0.55,
|
| 674 |
+
"grad_norm": 0.13623990945679543,
|
| 675 |
"learning_rate": 9.403848748301802e-05,
|
| 676 |
+
"loss": 0.5322,
|
| 677 |
"step": 92
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 0.55,
|
| 681 |
+
"grad_norm": 0.16411366119133766,
|
| 682 |
"learning_rate": 9.205498330282856e-05,
|
| 683 |
+
"loss": 0.5258,
|
| 684 |
"step": 93
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 0.56,
|
| 688 |
+
"grad_norm": 0.11646811294381437,
|
| 689 |
"learning_rate": 9.007462010919386e-05,
|
| 690 |
+
"loss": 0.4599,
|
| 691 |
"step": 94
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 0.56,
|
| 695 |
+
"grad_norm": 0.12610649830308363,
|
| 696 |
"learning_rate": 8.809818081980953e-05,
|
| 697 |
+
"loss": 0.4891,
|
| 698 |
"step": 95
|
| 699 |
},
|
| 700 |
{
|
| 701 |
"epoch": 0.57,
|
| 702 |
+
"grad_norm": 0.12324596375061997,
|
| 703 |
"learning_rate": 8.612644680109319e-05,
|
| 704 |
+
"loss": 0.4771,
|
| 705 |
"step": 96
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 0.58,
|
| 709 |
+
"grad_norm": 0.12390720748290898,
|
| 710 |
"learning_rate": 8.416019755927851e-05,
|
| 711 |
+
"loss": 0.4814,
|
| 712 |
"step": 97
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 0.58,
|
| 716 |
+
"grad_norm": 0.11156155557793686,
|
| 717 |
"learning_rate": 8.2200210432245e-05,
|
| 718 |
+
"loss": 0.5041,
|
| 719 |
"step": 98
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 0.59,
|
| 723 |
+
"grad_norm": 0.11235071151397323,
|
| 724 |
"learning_rate": 8.024726028220474e-05,
|
| 725 |
+
"loss": 0.5023,
|
| 726 |
"step": 99
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 0.59,
|
| 730 |
+
"grad_norm": 0.12382934807374943,
|
| 731 |
"learning_rate": 7.83021191893682e-05,
|
| 732 |
+
"loss": 0.4917,
|
| 733 |
"step": 100
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 0.6,
|
| 737 |
+
"grad_norm": 0.13051565063971357,
|
| 738 |
"learning_rate": 7.636555614670953e-05,
|
| 739 |
+
"loss": 0.4567,
|
| 740 |
"step": 101
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 0.61,
|
| 744 |
+
"grad_norm": 0.11975246976802223,
|
| 745 |
"learning_rate": 7.443833675595255e-05,
|
| 746 |
+
"loss": 0.4584,
|
| 747 |
"step": 102
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 0.61,
|
| 751 |
+
"grad_norm": 0.1390325061190883,
|
| 752 |
"learning_rate": 7.252122292489747e-05,
|
| 753 |
+
"loss": 0.4853,
|
| 754 |
"step": 103
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 0.62,
|
| 758 |
+
"grad_norm": 0.13092284300351015,
|
| 759 |
"learning_rate": 7.061497256620793e-05,
|
| 760 |
+
"loss": 0.4623,
|
| 761 |
"step": 104
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 0.62,
|
| 765 |
+
"grad_norm": 0.1291139691922856,
|
| 766 |
"learning_rate": 6.87203392977773e-05,
|
| 767 |
+
"loss": 0.4826,
|
| 768 |
"step": 105
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 0.63,
|
| 772 |
+
"grad_norm": 0.12630345566649634,
|
| 773 |
"learning_rate": 6.683807214479323e-05,
|
| 774 |
+
"loss": 0.4597,
|
| 775 |
"step": 106
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 0.64,
|
| 779 |
+
"grad_norm": 0.13061642512554117,
|
| 780 |
"learning_rate": 6.496891524361757e-05,
|
| 781 |
+
"loss": 0.4415,
|
| 782 |
"step": 107
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 0.64,
|
| 786 |
+
"grad_norm": 0.11414947835097713,
|
| 787 |
"learning_rate": 6.311360754759923e-05,
|
| 788 |
+
"loss": 0.4011,
|
| 789 |
"step": 108
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 0.65,
|
| 793 |
+
"grad_norm": 0.11453014913188615,
|
| 794 |
"learning_rate": 6.127288253493591e-05,
|
| 795 |
+
"loss": 0.5118,
|
| 796 |
"step": 109
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 0.65,
|
| 800 |
+
"grad_norm": 0.15000233747727326,
|
| 801 |
"learning_rate": 5.9447467918700614e-05,
|
| 802 |
+
"loss": 0.482,
|
| 803 |
"step": 110
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 0.66,
|
| 807 |
+
"grad_norm": 0.14402898946913723,
|
| 808 |
"learning_rate": 5.763808535914723e-05,
|
| 809 |
+
"loss": 0.4881,
|
| 810 |
"step": 111
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 0.67,
|
| 814 |
+
"grad_norm": 0.1371953423024685,
|
| 815 |
"learning_rate": 5.584545017840885e-05,
|
| 816 |
+
"loss": 0.5178,
|
| 817 |
"step": 112
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 0.67,
|
| 821 |
+
"grad_norm": 0.14105581149485616,
|
| 822 |
"learning_rate": 5.407027107770219e-05,
|
| 823 |
+
"loss": 0.5584,
|
| 824 |
"step": 113
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 0.68,
|
| 828 |
+
"grad_norm": 0.13646035299068404,
|
| 829 |
"learning_rate": 5.2313249857149414e-05,
|
| 830 |
+
"loss": 0.4535,
|
| 831 |
"step": 114
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 0.68,
|
| 835 |
+
"grad_norm": 0.14741390311838884,
|
| 836 |
"learning_rate": 5.0575081138327715e-05,
|
| 837 |
+
"loss": 0.5,
|
| 838 |
"step": 115
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 0.69,
|
| 842 |
+
"grad_norm": 0.12313483515418376,
|
| 843 |
"learning_rate": 4.885645208965779e-05,
|
| 844 |
+
"loss": 0.4982,
|
| 845 |
"step": 116
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 0.7,
|
| 849 |
+
"grad_norm": 0.13192008123654636,
|
| 850 |
"learning_rate": 4.715804215473809e-05,
|
| 851 |
+
"loss": 0.4698,
|
| 852 |
"step": 117
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 0.7,
|
| 856 |
+
"grad_norm": 0.11197839688620534,
|
| 857 |
"learning_rate": 4.548052278373327e-05,
|
| 858 |
+
"loss": 0.4737,
|
| 859 |
"step": 118
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 0.71,
|
| 863 |
+
"grad_norm": 0.12574619847347493,
|
| 864 |
"learning_rate": 4.382455716792291e-05,
|
| 865 |
+
"loss": 0.472,
|
| 866 |
"step": 119
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 0.71,
|
| 870 |
+
"grad_norm": 0.11785384512597769,
|
| 871 |
"learning_rate": 4.219079997751515e-05,
|
| 872 |
"loss": 0.4954,
|
| 873 |
"step": 120
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 0.72,
|
| 877 |
+
"grad_norm": 0.1371137498771671,
|
| 878 |
"learning_rate": 4.0579897102828966e-05,
|
| 879 |
+
"loss": 0.4645,
|
| 880 |
"step": 121
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 0.73,
|
| 884 |
+
"grad_norm": 0.1562115085397725,
|
| 885 |
"learning_rate": 3.899248539894757e-05,
|
| 886 |
+
"loss": 0.4798,
|
| 887 |
"step": 122
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 0.73,
|
| 891 |
+
"grad_norm": 0.14251919403809987,
|
| 892 |
"learning_rate": 3.7429192433944014e-05,
|
| 893 |
+
"loss": 0.4794,
|
| 894 |
"step": 123
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 0.74,
|
| 898 |
+
"grad_norm": 0.13567896269328303,
|
| 899 |
"learning_rate": 3.589063624077802e-05,
|
| 900 |
+
"loss": 0.4441,
|
| 901 |
"step": 124
|
| 902 |
},
|
| 903 |
{
|
| 904 |
"epoch": 0.74,
|
| 905 |
+
"grad_norm": 0.14154096161651117,
|
| 906 |
"learning_rate": 3.4377425072962465e-05,
|
| 907 |
+
"loss": 0.4583,
|
| 908 |
"step": 125
|
| 909 |
},
|
| 910 |
{
|
| 911 |
"epoch": 0.75,
|
| 912 |
+
"grad_norm": 0.12685010104316322,
|
| 913 |
"learning_rate": 3.289015716409631e-05,
|
| 914 |
+
"loss": 0.4445,
|
| 915 |
"step": 126
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 0.75,
|
| 919 |
+
"eval_loss": 0.9644750952720642,
|
| 920 |
+
"eval_runtime": 155.1633,
|
| 921 |
+
"eval_samples_per_second": 1.605,
|
| 922 |
+
"eval_steps_per_second": 0.806,
|
| 923 |
"step": 126
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 0.75,
|
| 927 |
+
"grad_norm": 0.127123263865781,
|
| 928 |
"learning_rate": 3.14294204913587e-05,
|
| 929 |
+
"loss": 0.4723,
|
| 930 |
"step": 127
|
| 931 |
},
|
| 932 |
{
|
| 933 |
"epoch": 0.76,
|
| 934 |
+
"grad_norm": 0.1220769041189061,
|
| 935 |
"learning_rate": 2.9995792543057478e-05,
|
| 936 |
+
"loss": 0.478,
|
| 937 |
"step": 128
|
| 938 |
},
|
| 939 |
{
|
| 940 |
"epoch": 0.77,
|
| 941 |
+
"grad_norm": 0.11180771873074949,
|
| 942 |
"learning_rate": 2.8589840090325027e-05,
|
| 943 |
+
"loss": 0.4909,
|
| 944 |
"step": 129
|
| 945 |
},
|
| 946 |
{
|
| 947 |
"epoch": 0.77,
|
| 948 |
+
"grad_norm": 0.13041544813383119,
|
| 949 |
"learning_rate": 2.7212118963050592e-05,
|
| 950 |
+
"loss": 0.4866,
|
| 951 |
"step": 130
|
| 952 |
},
|
| 953 |
{
|
| 954 |
"epoch": 0.78,
|
| 955 |
+
"grad_norm": 0.1405254912635375,
|
| 956 |
"learning_rate": 2.586317383013821e-05,
|
| 957 |
+
"loss": 0.4644,
|
| 958 |
"step": 131
|
| 959 |
},
|
| 960 |
{
|
| 961 |
"epoch": 0.78,
|
| 962 |
+
"grad_norm": 0.1344866421919383,
|
| 963 |
"learning_rate": 2.4543537984176978e-05,
|
| 964 |
+
"loss": 0.4456,
|
| 965 |
"step": 132
|
| 966 |
},
|
| 967 |
{
|
| 968 |
"epoch": 0.79,
|
| 969 |
+
"grad_norm": 0.11186073290835527,
|
| 970 |
"learning_rate": 2.325373313060919e-05,
|
| 971 |
+
"loss": 0.4994,
|
| 972 |
"step": 133
|
| 973 |
},
|
| 974 |
{
|
| 975 |
"epoch": 0.8,
|
| 976 |
+
"grad_norm": 0.11996543804410408,
|
| 977 |
"learning_rate": 2.19942691814788e-05,
|
| 978 |
+
"loss": 0.5109,
|
| 979 |
"step": 134
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 0.8,
|
| 983 |
+
"grad_norm": 0.1213151845276305,
|
| 984 |
"learning_rate": 2.076564405384258e-05,
|
| 985 |
+
"loss": 0.4519,
|
| 986 |
"step": 135
|
| 987 |
},
|
| 988 |
{
|
| 989 |
"epoch": 0.81,
|
| 990 |
+
"grad_norm": 0.13448384445509487,
|
| 991 |
"learning_rate": 1.9568343472923524e-05,
|
| 992 |
+
"loss": 0.5102,
|
| 993 |
"step": 136
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 0.81,
|
| 997 |
+
"grad_norm": 0.12446550745886309,
|
| 998 |
"learning_rate": 1.840284078008393e-05,
|
| 999 |
+
"loss": 0.5064,
|
| 1000 |
"step": 137
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
"epoch": 0.82,
|
| 1004 |
+
"grad_norm": 0.14018404224721664,
|
| 1005 |
"learning_rate": 1.7269596745694295e-05,
|
| 1006 |
+
"loss": 0.4805,
|
| 1007 |
"step": 138
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 0.83,
|
| 1011 |
+
"grad_norm": 0.1218286028602253,
|
| 1012 |
"learning_rate": 1.616905938697234e-05,
|
| 1013 |
+
"loss": 0.4388,
|
| 1014 |
"step": 139
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
"epoch": 0.83,
|
| 1018 |
+
"grad_norm": 0.1293899895212843,
|
| 1019 |
"learning_rate": 1.5101663790863596e-05,
|
| 1020 |
+
"loss": 0.4935,
|
| 1021 |
"step": 140
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
"epoch": 0.84,
|
| 1025 |
+
"grad_norm": 0.1338746361041889,
|
| 1026 |
"learning_rate": 1.4067831942033904e-05,
|
| 1027 |
+
"loss": 0.4892,
|
| 1028 |
"step": 141
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 0.84,
|
| 1032 |
+
"grad_norm": 0.12837018186336283,
|
| 1033 |
"learning_rate": 1.3067972556041752e-05,
|
| 1034 |
+
"loss": 0.4566,
|
| 1035 |
"step": 142
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
"epoch": 0.85,
|
| 1039 |
+
"grad_norm": 0.1359601843202103,
|
| 1040 |
"learning_rate": 1.210248091775663e-05,
|
| 1041 |
+
"loss": 0.5277,
|
| 1042 |
"step": 143
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 0.86,
|
| 1046 |
+
"grad_norm": 0.14494892289523706,
|
| 1047 |
"learning_rate": 1.1171738725086833e-05,
|
| 1048 |
+
"loss": 0.4419,
|
| 1049 |
"step": 144
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 0.86,
|
| 1053 |
+
"grad_norm": 0.13277965728610747,
|
| 1054 |
"learning_rate": 1.0276113938078769e-05,
|
| 1055 |
+
"loss": 0.4662,
|
| 1056 |
"step": 145
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 0.87,
|
| 1060 |
+
"grad_norm": 0.12880953859990588,
|
| 1061 |
"learning_rate": 9.415960633447674e-06,
|
| 1062 |
"loss": 0.4452,
|
| 1063 |
"step": 146
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 0.87,
|
| 1067 |
+
"grad_norm": 0.13140413424345554,
|
| 1068 |
"learning_rate": 8.59161886459654e-06,
|
| 1069 |
+
"loss": 0.485,
|
| 1070 |
"step": 147
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
"epoch": 0.88,
|
| 1074 |
+
"grad_norm": 0.1319983788489183,
|
| 1075 |
"learning_rate": 7.803414527179343e-06,
|
| 1076 |
+
"loss": 0.4601,
|
| 1077 |
"step": 148
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
"epoch": 0.89,
|
| 1081 |
+
"grad_norm": 0.12708833536742306,
|
| 1082 |
"learning_rate": 7.051659230261298e-06,
|
| 1083 |
+
"loss": 0.498,
|
| 1084 |
"step": 149
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 0.89,
|
| 1088 |
+
"grad_norm": 0.13606238849102634,
|
| 1089 |
"learning_rate": 6.336650173127223e-06,
|
| 1090 |
+
"loss": 0.504,
|
| 1091 |
"step": 150
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 0.9,
|
| 1095 |
+
"grad_norm": 0.1295595089005434,
|
| 1096 |
"learning_rate": 5.658670027786561e-06,
|
| 1097 |
+
"loss": 0.4704,
|
| 1098 |
"step": 151
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
"epoch": 0.9,
|
| 1102 |
+
"grad_norm": 0.1426317101207339,
|
| 1103 |
"learning_rate": 5.017986827221733e-06,
|
| 1104 |
+
"loss": 0.4979,
|
| 1105 |
"step": 152
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 0.91,
|
| 1109 |
+
"grad_norm": 0.14306220741750347,
|
| 1110 |
"learning_rate": 4.4148538594239174e-06,
|
| 1111 |
+
"loss": 0.5164,
|
| 1112 |
"step": 153
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 0.92,
|
| 1116 |
+
"grad_norm": 0.11966060296362775,
|
| 1117 |
"learning_rate": 3.849509567257959e-06,
|
| 1118 |
+
"loss": 0.508,
|
| 1119 |
"step": 154
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
"epoch": 0.92,
|
| 1123 |
+
"grad_norm": 0.13410648920269255,
|
| 1124 |
"learning_rate": 3.3221774541962845e-06,
|
| 1125 |
+
"loss": 0.4647,
|
| 1126 |
"step": 155
|
| 1127 |
},
|
| 1128 |
{
|
| 1129 |
"epoch": 0.93,
|
| 1130 |
+
"grad_norm": 0.14279456888939884,
|
| 1131 |
"learning_rate": 2.8330659959589946e-06,
|
| 1132 |
+
"loss": 0.4446,
|
| 1133 |
"step": 156
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
"epoch": 0.93,
|
| 1137 |
+
"grad_norm": 0.12867690404851015,
|
| 1138 |
"learning_rate": 2.3823685580949273e-06,
|
| 1139 |
+
"loss": 0.453,
|
| 1140 |
"step": 157
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
"epoch": 0.94,
|
| 1144 |
+
"grad_norm": 0.12934543469906262,
|
| 1145 |
"learning_rate": 1.9702633195363917e-06,
|
| 1146 |
+
"loss": 0.4782,
|
| 1147 |
"step": 158
|
| 1148 |
},
|
| 1149 |
{
|
| 1150 |
"epoch": 0.95,
|
| 1151 |
+
"grad_norm": 0.12938497975922164,
|
| 1152 |
"learning_rate": 1.5969132021579347e-06,
|
| 1153 |
+
"loss": 0.5004,
|
| 1154 |
"step": 159
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
"epoch": 0.95,
|
| 1158 |
+
"grad_norm": 0.12309207540187947,
|
| 1159 |
"learning_rate": 1.2624658063666639e-06,
|
| 1160 |
+
"loss": 0.5235,
|
| 1161 |
"step": 160
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
"epoch": 0.96,
|
| 1165 |
+
"grad_norm": 0.11548916080113472,
|
| 1166 |
"learning_rate": 9.670533527498137e-07,
|
| 1167 |
+
"loss": 0.4436,
|
| 1168 |
"step": 161
|
| 1169 |
},
|
| 1170 |
{
|
| 1171 |
"epoch": 0.96,
|
| 1172 |
+
"grad_norm": 0.12421128547481033,
|
| 1173 |
"learning_rate": 7.10792629802659e-07,
|
| 1174 |
+
"loss": 0.493,
|
| 1175 |
"step": 162
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
"epoch": 0.97,
|
| 1179 |
+
"grad_norm": 0.12845932182275463,
|
| 1180 |
"learning_rate": 4.937849477572587e-07,
|
| 1181 |
+
"loss": 0.4322,
|
| 1182 |
"step": 163
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
"epoch": 0.97,
|
| 1186 |
+
"grad_norm": 0.12044022703621939,
|
| 1187 |
"learning_rate": 3.161160985304168e-07,
|
| 1188 |
+
"loss": 0.4261,
|
| 1189 |
"step": 164
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
"epoch": 0.98,
|
| 1193 |
+
"grad_norm": 0.12854436448690823,
|
| 1194 |
"learning_rate": 1.7785632180670198e-07,
|
| 1195 |
"loss": 0.436,
|
| 1196 |
"step": 165
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
"epoch": 0.99,
|
| 1200 |
+
"grad_norm": 0.12592400334168094,
|
| 1201 |
"learning_rate": 7.906027726981568e-08,
|
| 1202 |
+
"loss": 0.5008,
|
| 1203 |
"step": 166
|
| 1204 |
},
|
| 1205 |
{
|
| 1206 |
"epoch": 0.99,
|
| 1207 |
+
"grad_norm": 0.13604136642444342,
|
| 1208 |
"learning_rate": 1.976702299344435e-08,
|
| 1209 |
+
"loss": 0.5117,
|
| 1210 |
"step": 167
|
| 1211 |
},
|
| 1212 |
{
|
| 1213 |
"epoch": 1.0,
|
| 1214 |
+
"grad_norm": 0.12600479253243235,
|
| 1215 |
"learning_rate": 0.0,
|
| 1216 |
+
"loss": 0.4609,
|
| 1217 |
"step": 168
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 1.0,
|
| 1221 |
+
"eval_loss": 0.9640664458274841,
|
| 1222 |
+
"eval_runtime": 154.1049,
|
| 1223 |
+
"eval_samples_per_second": 1.616,
|
| 1224 |
+
"eval_steps_per_second": 0.811,
|
| 1225 |
"step": 168
|
| 1226 |
}
|
| 1227 |
],
|
checkpoint-168/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6968
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3baf97ea9bcc6b73a05f59bedb019ebf1656742a3d0668caac6a61c4da440046
|
| 3 |
size 6968
|
config.json
CHANGED
|
@@ -37,13 +37,28 @@
|
|
| 37 |
"num_key_value_heads": 8,
|
| 38 |
"output_router_logits": false,
|
| 39 |
"pad_token_id": 0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
"rms_norm_eps": 1e-06,
|
| 41 |
"router_aux_loss_coef": 0.001,
|
| 42 |
"sliding_window": null,
|
| 43 |
"tie_word_embeddings": false,
|
| 44 |
-
"torch_dtype": "
|
| 45 |
"transformers_version": "4.40.0.dev0",
|
| 46 |
-
"use_cache":
|
| 47 |
"use_mamba_kernels": true,
|
| 48 |
"vocab_size": 65536
|
| 49 |
}
|
|
|
|
| 37 |
"num_key_value_heads": 8,
|
| 38 |
"output_router_logits": false,
|
| 39 |
"pad_token_id": 0,
|
| 40 |
+
"quantization_config": {
|
| 41 |
+
"_load_in_4bit": true,
|
| 42 |
+
"_load_in_8bit": false,
|
| 43 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 44 |
+
"bnb_4bit_quant_storage": "bfloat16",
|
| 45 |
+
"bnb_4bit_quant_type": "nf4",
|
| 46 |
+
"bnb_4bit_use_double_quant": true,
|
| 47 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 48 |
+
"llm_int8_has_fp16_weight": false,
|
| 49 |
+
"llm_int8_skip_modules": null,
|
| 50 |
+
"llm_int8_threshold": 6.0,
|
| 51 |
+
"load_in_4bit": true,
|
| 52 |
+
"load_in_8bit": false,
|
| 53 |
+
"quant_method": "bitsandbytes"
|
| 54 |
+
},
|
| 55 |
"rms_norm_eps": 1e-06,
|
| 56 |
"router_aux_loss_coef": 0.001,
|
| 57 |
"sliding_window": null,
|
| 58 |
"tie_word_embeddings": false,
|
| 59 |
+
"torch_dtype": "bfloat16",
|
| 60 |
"transformers_version": "4.40.0.dev0",
|
| 61 |
+
"use_cache": false,
|
| 62 |
"use_mamba_kernels": true,
|
| 63 |
"vocab_size": 65536
|
| 64 |
}
|
tokenizer_config.json
CHANGED
|
@@ -36,6 +36,7 @@
|
|
| 36 |
}
|
| 37 |
},
|
| 38 |
"bos_token": "<|startoftext|>",
|
|
|
|
| 39 |
"clean_up_tokenization_spaces": false,
|
| 40 |
"eos_token": "<|endoftext|>",
|
| 41 |
"model_max_length": 1000000000000000019884624838656,
|
|
|
|
| 36 |
}
|
| 37 |
},
|
| 38 |
"bos_token": "<|startoftext|>",
|
| 39 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 40 |
"clean_up_tokenization_spaces": false,
|
| 41 |
"eos_token": "<|endoftext|>",
|
| 42 |
"model_max_length": 1000000000000000019884624838656,
|