Model save
Browse files
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-7b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma7b-summarize-gpt4o-30k
|
| 14 |
results: []
|
|
@@ -17,12 +16,12 @@ model-index:
|
|
| 17 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
|
| 20 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/
|
| 21 |
# gemma7b-summarize-gpt4o-30k
|
| 22 |
|
| 23 |
-
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the
|
| 24 |
It achieves the following results on the evaluation set:
|
| 25 |
-
- Loss: 2.
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
@@ -53,13 +52,17 @@ The following hyperparameters were used during training:
|
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: cosine
|
| 55 |
- lr_scheduler_warmup_ratio: 0.1
|
| 56 |
-
- num_epochs:
|
| 57 |
|
| 58 |
### Training results
|
| 59 |
|
| 60 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 61 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 62 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
|
| 64 |
|
| 65 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-7b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma7b-summarize-gpt4o-30k
|
| 13 |
results: []
|
|
|
|
| 16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
|
| 19 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/ddvw2m8z)
|
| 20 |
# gemma7b-summarize-gpt4o-30k
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 2.3811
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
| 54 |
- lr_scheduler_warmup_ratio: 0.1
|
| 55 |
+
- num_epochs: 5
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
+
| 0.9712 | 1.0 | 137 | 2.3077 |
|
| 62 |
+
| 0.8675 | 2.0 | 274 | 2.2479 |
|
| 63 |
+
| 0.7623 | 3.0 | 411 | 2.2756 |
|
| 64 |
+
| 0.709 | 4.0 | 548 | 2.3417 |
|
| 65 |
+
| 0.6601 | 5.0 | 685 | 2.3811 |
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 50056096
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9dead167941940b2975197a1f402c334b3606ea3d8e60a06c79db262f8105d00
|
| 3 |
size 50056096
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 4.87,
|
| 7 |
-
"eval_steps_per_second": 1.948,
|
| 8 |
-
"total_flos": 2.0945562398778982e+17,
|
| 9 |
-
"train_loss": 4.772963228887015,
|
| 10 |
-
"train_runtime": 1080.9502,
|
| 11 |
"train_samples": 29787,
|
| 12 |
-
"train_samples_per_second": 2.
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.0472781231601746e+18,
|
| 4 |
+
"train_loss": 2.151051264783762,
|
| 5 |
+
"train_runtime": 5341.9856,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 29787,
|
| 7 |
+
"train_samples_per_second": 2.052,
|
| 8 |
+
"train_steps_per_second": 0.128
|
| 9 |
}
|
runs/May19_11-54-28_deep-diver-main-lucky-mouse-1-0-0/events.out.tfevents.1716134219.deep-diver-main-lucky-mouse-1-0-0.385.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b9bd7ceaaf98ed46022543b9806d0f0f89b6e1bbda57604b1783852001cedbc
|
| 3 |
+
size 36313
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 29787,
|
| 7 |
-
"train_samples_per_second": 2.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.0472781231601746e+18,
|
| 4 |
+
"train_loss": 2.151051264783762,
|
| 5 |
+
"train_runtime": 5341.9856,
|
| 6 |
"train_samples": 29787,
|
| 7 |
+
"train_samples_per_second": 2.052,
|
| 8 |
+
"train_steps_per_second": 0.128
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,221 +11,1023 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.0072992700729927005,
|
| 13 |
"grad_norm": 708.0,
|
| 14 |
-
"learning_rate":
|
| 15 |
"loss": 56.8346,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.0364963503649635,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.072992700729927,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.10948905109489052,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.145985401459854,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.18248175182481752,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.21897810218978103,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.25547445255474455,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.291970802919708,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.3284671532846715,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.36496350364963503,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.40145985401459855,
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.43795620437956206,
|
| 97 |
-
"grad_norm": 3.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.4744525547445255,
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.5109489051094891,
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.5474452554744526,
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.583941605839416,
|
| 125 |
-
"grad_norm": 2.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.6204379562043796,
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.656934306569343,
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.6934306569343066,
|
| 146 |
-
"grad_norm": 1.
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.7299270072992701,
|
| 153 |
-
"grad_norm": 1.
|
| 154 |
-
"learning_rate":
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.7664233576642335,
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate":
|
| 162 |
-
"loss":
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.8029197080291971,
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss":
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.8394160583941606,
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.8759124087591241,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.9124087591240876,
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.948905109489051,
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.9854014598540146,
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate":
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 1.0,
|
| 209 |
-
"eval_loss": 2.
|
| 210 |
-
"eval_runtime":
|
| 211 |
-
"eval_samples_per_second":
|
| 212 |
-
"eval_steps_per_second":
|
| 213 |
"step": 137
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 1.
|
| 217 |
-
"
|
| 218 |
-
"
|
| 219 |
-
"
|
| 220 |
-
"
|
| 221 |
-
|
| 222 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 223 |
}
|
| 224 |
],
|
| 225 |
"logging_steps": 5,
|
| 226 |
-
"max_steps":
|
| 227 |
"num_input_tokens_seen": 0,
|
| 228 |
-
"num_train_epochs":
|
| 229 |
"save_steps": 100,
|
| 230 |
"stateful_callbacks": {
|
| 231 |
"TrainerControl": {
|
|
@@ -239,7 +1041,7 @@
|
|
| 239 |
"attributes": {}
|
| 240 |
}
|
| 241 |
},
|
| 242 |
-
"total_flos":
|
| 243 |
"train_batch_size": 4,
|
| 244 |
"trial_name": null,
|
| 245 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 685,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.0072992700729927005,
|
| 13 |
"grad_norm": 708.0,
|
| 14 |
+
"learning_rate": 2.898550724637681e-06,
|
| 15 |
"loss": 56.8346,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.0364963503649635,
|
| 20 |
+
"grad_norm": 604.0,
|
| 21 |
+
"learning_rate": 1.4492753623188407e-05,
|
| 22 |
+
"loss": 52.9742,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.072992700729927,
|
| 27 |
+
"grad_norm": 340.0,
|
| 28 |
+
"learning_rate": 2.8985507246376814e-05,
|
| 29 |
+
"loss": 39.0746,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.10948905109489052,
|
| 34 |
+
"grad_norm": 40.25,
|
| 35 |
+
"learning_rate": 4.347826086956522e-05,
|
| 36 |
+
"loss": 20.8099,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.145985401459854,
|
| 41 |
+
"grad_norm": 25.5,
|
| 42 |
+
"learning_rate": 5.797101449275363e-05,
|
| 43 |
+
"loss": 17.6144,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.18248175182481752,
|
| 48 |
+
"grad_norm": 7.78125,
|
| 49 |
+
"learning_rate": 7.246376811594203e-05,
|
| 50 |
+
"loss": 15.3803,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.21897810218978103,
|
| 55 |
+
"grad_norm": 6.40625,
|
| 56 |
+
"learning_rate": 8.695652173913044e-05,
|
| 57 |
+
"loss": 14.0798,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.25547445255474455,
|
| 62 |
+
"grad_norm": 13.4375,
|
| 63 |
+
"learning_rate": 0.00010144927536231885,
|
| 64 |
+
"loss": 13.4032,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.291970802919708,
|
| 69 |
+
"grad_norm": 41.0,
|
| 70 |
+
"learning_rate": 0.00011594202898550725,
|
| 71 |
+
"loss": 10.8827,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.3284671532846715,
|
| 76 |
+
"grad_norm": 13.1875,
|
| 77 |
+
"learning_rate": 0.00013043478260869567,
|
| 78 |
+
"loss": 4.5915,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.36496350364963503,
|
| 83 |
+
"grad_norm": 4.09375,
|
| 84 |
+
"learning_rate": 0.00014492753623188405,
|
| 85 |
+
"loss": 1.9,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.40145985401459855,
|
| 90 |
+
"grad_norm": 2.28125,
|
| 91 |
+
"learning_rate": 0.00015942028985507247,
|
| 92 |
+
"loss": 1.6474,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.43795620437956206,
|
| 97 |
+
"grad_norm": 3.5,
|
| 98 |
+
"learning_rate": 0.00017391304347826088,
|
| 99 |
+
"loss": 1.477,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.4744525547445255,
|
| 104 |
+
"grad_norm": 2.28125,
|
| 105 |
+
"learning_rate": 0.00018840579710144927,
|
| 106 |
+
"loss": 1.3309,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.5109489051094891,
|
| 111 |
+
"grad_norm": 1.6171875,
|
| 112 |
+
"learning_rate": 0.00019999869950890106,
|
| 113 |
+
"loss": 1.2538,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.5474452554744526,
|
| 118 |
+
"grad_norm": 5.9375,
|
| 119 |
+
"learning_rate": 0.0001999531858720213,
|
| 120 |
+
"loss": 1.224,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.583941605839416,
|
| 125 |
+
"grad_norm": 2.25,
|
| 126 |
+
"learning_rate": 0.00019984268150178167,
|
| 127 |
+
"loss": 1.1823,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.6204379562043796,
|
| 132 |
+
"grad_norm": 2.078125,
|
| 133 |
+
"learning_rate": 0.00019966725824941932,
|
| 134 |
+
"loss": 1.1279,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.656934306569343,
|
| 139 |
+
"grad_norm": 3.0625,
|
| 140 |
+
"learning_rate": 0.00019942703017718975,
|
| 141 |
+
"loss": 1.127,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.6934306569343066,
|
| 146 |
+
"grad_norm": 1.75,
|
| 147 |
+
"learning_rate": 0.000199122153484202,
|
| 148 |
+
"loss": 1.1284,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.7299270072992701,
|
| 153 |
+
"grad_norm": 1.5625,
|
| 154 |
+
"learning_rate": 0.00019875282640485645,
|
| 155 |
+
"loss": 1.0566,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.7664233576642335,
|
| 160 |
+
"grad_norm": 4.53125,
|
| 161 |
+
"learning_rate": 0.0001983192890799503,
|
| 162 |
+
"loss": 1.0361,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.8029197080291971,
|
| 167 |
+
"grad_norm": 2.5,
|
| 168 |
+
"learning_rate": 0.0001978218234005352,
|
| 169 |
+
"loss": 1.0371,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.8394160583941606,
|
| 174 |
+
"grad_norm": 1.890625,
|
| 175 |
+
"learning_rate": 0.00019726075282462845,
|
| 176 |
+
"loss": 1.0235,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.8759124087591241,
|
| 181 |
+
"grad_norm": 0.67578125,
|
| 182 |
+
"learning_rate": 0.00019663644216689683,
|
| 183 |
+
"loss": 0.996,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.9124087591240876,
|
| 188 |
+
"grad_norm": 1.2421875,
|
| 189 |
+
"learning_rate": 0.00019594929736144976,
|
| 190 |
+
"loss": 0.9734,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.948905109489051,
|
| 195 |
+
"grad_norm": 1.5625,
|
| 196 |
+
"learning_rate": 0.00019519976519789616,
|
| 197 |
+
"loss": 0.978,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.9854014598540146,
|
| 202 |
+
"grad_norm": 0.95703125,
|
| 203 |
+
"learning_rate": 0.00019438833303083678,
|
| 204 |
+
"loss": 0.9712,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 1.0,
|
| 209 |
+
"eval_loss": 2.307734489440918,
|
| 210 |
+
"eval_runtime": 0.9962,
|
| 211 |
+
"eval_samples_per_second": 5.019,
|
| 212 |
+
"eval_steps_per_second": 2.008,
|
| 213 |
"step": 137
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 1.0218978102189782,
|
| 217 |
+
"grad_norm": 2.125,
|
| 218 |
+
"learning_rate": 0.00019351552846298025,
|
| 219 |
+
"loss": 0.9374,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.0583941605839415,
|
| 224 |
+
"grad_norm": 2.265625,
|
| 225 |
+
"learning_rate": 0.0001925819190020898,
|
| 226 |
+
"loss": 0.9173,
|
| 227 |
+
"step": 145
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.094890510948905,
|
| 231 |
+
"grad_norm": 0.828125,
|
| 232 |
+
"learning_rate": 0.00019158811169198313,
|
| 233 |
+
"loss": 0.8916,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.1313868613138687,
|
| 238 |
+
"grad_norm": 1.0703125,
|
| 239 |
+
"learning_rate": 0.0001905347527178252,
|
| 240 |
+
"loss": 0.9418,
|
| 241 |
+
"step": 155
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.167883211678832,
|
| 245 |
+
"grad_norm": 0.9140625,
|
| 246 |
+
"learning_rate": 0.00018942252698597113,
|
| 247 |
+
"loss": 0.9054,
|
| 248 |
+
"step": 160
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.2043795620437956,
|
| 252 |
+
"grad_norm": 2.0625,
|
| 253 |
+
"learning_rate": 0.00018825215767863214,
|
| 254 |
+
"loss": 0.9039,
|
| 255 |
+
"step": 165
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.2408759124087592,
|
| 259 |
+
"grad_norm": 1.5859375,
|
| 260 |
+
"learning_rate": 0.00018702440578365387,
|
| 261 |
+
"loss": 0.9146,
|
| 262 |
+
"step": 170
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.2773722627737225,
|
| 266 |
+
"grad_norm": 1.3515625,
|
| 267 |
+
"learning_rate": 0.00018574006959971333,
|
| 268 |
+
"loss": 0.8896,
|
| 269 |
+
"step": 175
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.313868613138686,
|
| 273 |
+
"grad_norm": 2.09375,
|
| 274 |
+
"learning_rate": 0.00018439998421725554,
|
| 275 |
+
"loss": 0.8947,
|
| 276 |
+
"step": 180
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.3503649635036497,
|
| 280 |
+
"grad_norm": 0.80078125,
|
| 281 |
+
"learning_rate": 0.00018300502097550806,
|
| 282 |
+
"loss": 0.881,
|
| 283 |
+
"step": 185
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.3868613138686132,
|
| 287 |
+
"grad_norm": 0.80078125,
|
| 288 |
+
"learning_rate": 0.00018155608689592604,
|
| 289 |
+
"loss": 0.8906,
|
| 290 |
+
"step": 190
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.4233576642335766,
|
| 294 |
+
"grad_norm": 0.80859375,
|
| 295 |
+
"learning_rate": 0.00018005412409243606,
|
| 296 |
+
"loss": 0.8939,
|
| 297 |
+
"step": 195
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.4598540145985401,
|
| 301 |
+
"grad_norm": 1.0234375,
|
| 302 |
+
"learning_rate": 0.0001785001091588628,
|
| 303 |
+
"loss": 0.9016,
|
| 304 |
+
"step": 200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.4963503649635037,
|
| 308 |
+
"grad_norm": 0.70703125,
|
| 309 |
+
"learning_rate": 0.0001768950525339362,
|
| 310 |
+
"loss": 0.8943,
|
| 311 |
+
"step": 205
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.5328467153284673,
|
| 315 |
+
"grad_norm": 1.2109375,
|
| 316 |
+
"learning_rate": 0.00017523999784429238,
|
| 317 |
+
"loss": 0.8614,
|
| 318 |
+
"step": 210
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.5693430656934306,
|
| 322 |
+
"grad_norm": 0.7734375,
|
| 323 |
+
"learning_rate": 0.00017353602122589527,
|
| 324 |
+
"loss": 0.8788,
|
| 325 |
+
"step": 215
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.6058394160583942,
|
| 329 |
+
"grad_norm": 0.82421875,
|
| 330 |
+
"learning_rate": 0.0001717842306243205,
|
| 331 |
+
"loss": 0.8833,
|
| 332 |
+
"step": 220
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.6423357664233578,
|
| 336 |
+
"grad_norm": 0.84765625,
|
| 337 |
+
"learning_rate": 0.00016998576507435618,
|
| 338 |
+
"loss": 0.8713,
|
| 339 |
+
"step": 225
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.6788321167883211,
|
| 343 |
+
"grad_norm": 1.234375,
|
| 344 |
+
"learning_rate": 0.00016814179395938913,
|
| 345 |
+
"loss": 0.8661,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.7153284671532847,
|
| 350 |
+
"grad_norm": 0.91015625,
|
| 351 |
+
"learning_rate": 0.00016625351625105796,
|
| 352 |
+
"loss": 0.8413,
|
| 353 |
+
"step": 235
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.7518248175182483,
|
| 357 |
+
"grad_norm": 0.63671875,
|
| 358 |
+
"learning_rate": 0.0001643221597296679,
|
| 359 |
+
"loss": 0.8741,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.7883211678832116,
|
| 364 |
+
"grad_norm": 0.73046875,
|
| 365 |
+
"learning_rate": 0.00016234898018587337,
|
| 366 |
+
"loss": 0.8744,
|
| 367 |
+
"step": 245
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.8248175182481752,
|
| 371 |
+
"grad_norm": 0.671875,
|
| 372 |
+
"learning_rate": 0.00016033526060414842,
|
| 373 |
+
"loss": 0.8517,
|
| 374 |
+
"step": 250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.8613138686131387,
|
| 378 |
+
"grad_norm": 1.0234375,
|
| 379 |
+
"learning_rate": 0.00015828231032857503,
|
| 380 |
+
"loss": 0.8899,
|
| 381 |
+
"step": 255
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.897810218978102,
|
| 385 |
+
"grad_norm": 0.66796875,
|
| 386 |
+
"learning_rate": 0.00015619146421149232,
|
| 387 |
+
"loss": 0.8537,
|
| 388 |
+
"step": 260
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.9343065693430657,
|
| 392 |
+
"grad_norm": 0.7109375,
|
| 393 |
+
"learning_rate": 0.00015406408174555976,
|
| 394 |
+
"loss": 0.8329,
|
| 395 |
+
"step": 265
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.9708029197080292,
|
| 399 |
+
"grad_norm": 0.71875,
|
| 400 |
+
"learning_rate": 0.00015190154617979938,
|
| 401 |
+
"loss": 0.8675,
|
| 402 |
+
"step": 270
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 2.0,
|
| 406 |
+
"eval_loss": 2.247941017150879,
|
| 407 |
+
"eval_runtime": 0.9979,
|
| 408 |
+
"eval_samples_per_second": 5.01,
|
| 409 |
+
"eval_steps_per_second": 2.004,
|
| 410 |
+
"step": 274
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.0072992700729926,
|
| 414 |
+
"grad_norm": 0.80859375,
|
| 415 |
+
"learning_rate": 0.00014970526362019079,
|
| 416 |
+
"loss": 0.8435,
|
| 417 |
+
"step": 275
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.0437956204379564,
|
| 421 |
+
"grad_norm": 1.515625,
|
| 422 |
+
"learning_rate": 0.00014747666211540459,
|
| 423 |
+
"loss": 0.7774,
|
| 424 |
+
"step": 280
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.0802919708029197,
|
| 428 |
+
"grad_norm": 1.0859375,
|
| 429 |
+
"learning_rate": 0.00014521719072826858,
|
| 430 |
+
"loss": 0.79,
|
| 431 |
+
"step": 285
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.116788321167883,
|
| 435 |
+
"grad_norm": 0.498046875,
|
| 436 |
+
"learning_rate": 0.00014292831859356997,
|
| 437 |
+
"loss": 0.7929,
|
| 438 |
+
"step": 290
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.153284671532847,
|
| 442 |
+
"grad_norm": 1.59375,
|
| 443 |
+
"learning_rate": 0.00014061153396280674,
|
| 444 |
+
"loss": 0.8032,
|
| 445 |
+
"step": 295
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.18978102189781,
|
| 449 |
+
"grad_norm": 0.83203125,
|
| 450 |
+
"learning_rate": 0.000138268343236509,
|
| 451 |
+
"loss": 0.7932,
|
| 452 |
+
"step": 300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.2262773722627736,
|
| 456 |
+
"grad_norm": 0.734375,
|
| 457 |
+
"learning_rate": 0.00013590026998475986,
|
| 458 |
+
"loss": 0.7657,
|
| 459 |
+
"step": 305
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.2627737226277373,
|
| 463 |
+
"grad_norm": 0.609375,
|
| 464 |
+
"learning_rate": 0.0001335088539565523,
|
| 465 |
+
"loss": 0.783,
|
| 466 |
+
"step": 310
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.2992700729927007,
|
| 470 |
+
"grad_norm": 0.71484375,
|
| 471 |
+
"learning_rate": 0.00013109565007862596,
|
| 472 |
+
"loss": 0.7755,
|
| 473 |
+
"step": 315
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.335766423357664,
|
| 477 |
+
"grad_norm": 0.609375,
|
| 478 |
+
"learning_rate": 0.0001286622274444361,
|
| 479 |
+
"loss": 0.7723,
|
| 480 |
+
"step": 320
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.372262773722628,
|
| 484 |
+
"grad_norm": 1.3359375,
|
| 485 |
+
"learning_rate": 0.00012621016829391022,
|
| 486 |
+
"loss": 0.7739,
|
| 487 |
+
"step": 325
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.408759124087591,
|
| 491 |
+
"grad_norm": 1.1328125,
|
| 492 |
+
"learning_rate": 0.00012374106698465732,
|
| 493 |
+
"loss": 0.7821,
|
| 494 |
+
"step": 330
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.445255474452555,
|
| 498 |
+
"grad_norm": 0.91015625,
|
| 499 |
+
"learning_rate": 0.00012125652895529766,
|
| 500 |
+
"loss": 0.7852,
|
| 501 |
+
"step": 335
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.4817518248175183,
|
| 505 |
+
"grad_norm": 0.74609375,
|
| 506 |
+
"learning_rate": 0.00011875816968158815,
|
| 507 |
+
"loss": 0.7792,
|
| 508 |
+
"step": 340
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.5182481751824817,
|
| 512 |
+
"grad_norm": 0.625,
|
| 513 |
+
"learning_rate": 0.00011624761362602061,
|
| 514 |
+
"loss": 0.7799,
|
| 515 |
+
"step": 345
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.554744525547445,
|
| 519 |
+
"grad_norm": 0.81640625,
|
| 520 |
+
"learning_rate": 0.00011372649318157749,
|
| 521 |
+
"loss": 0.7914,
|
| 522 |
+
"step": 350
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.591240875912409,
|
| 526 |
+
"grad_norm": 0.80078125,
|
| 527 |
+
"learning_rate": 0.00011119644761033078,
|
| 528 |
+
"loss": 0.7847,
|
| 529 |
+
"step": 355
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.627737226277372,
|
| 533 |
+
"grad_norm": 0.984375,
|
| 534 |
+
"learning_rate": 0.0001086591219775746,
|
| 535 |
+
"loss": 0.8049,
|
| 536 |
+
"step": 360
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.664233576642336,
|
| 540 |
+
"grad_norm": 0.81640625,
|
| 541 |
+
"learning_rate": 0.00010611616608218429,
|
| 542 |
+
"loss": 0.7865,
|
| 543 |
+
"step": 365
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.7007299270072993,
|
| 547 |
+
"grad_norm": 0.51953125,
|
| 548 |
+
"learning_rate": 0.00010356923338389806,
|
| 549 |
+
"loss": 0.7908,
|
| 550 |
+
"step": 370
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.7372262773722627,
|
| 554 |
+
"grad_norm": 0.53125,
|
| 555 |
+
"learning_rate": 0.00010101997992821797,
|
| 556 |
+
"loss": 0.7925,
|
| 557 |
+
"step": 375
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.7737226277372264,
|
| 561 |
+
"grad_norm": 0.49609375,
|
| 562 |
+
"learning_rate": 9.847006326962974e-05,
|
| 563 |
+
"loss": 0.799,
|
| 564 |
+
"step": 380
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.81021897810219,
|
| 568 |
+
"grad_norm": 0.51171875,
|
| 569 |
+
"learning_rate": 9.592114139384145e-05,
|
| 570 |
+
"loss": 0.7832,
|
| 571 |
+
"step": 385
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.846715328467153,
|
| 575 |
+
"grad_norm": 0.7109375,
|
| 576 |
+
"learning_rate": 9.337487163974164e-05,
|
| 577 |
+
"loss": 0.7796,
|
| 578 |
+
"step": 390
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.883211678832117,
|
| 582 |
+
"grad_norm": 0.6328125,
|
| 583 |
+
"learning_rate": 9.083290962177828e-05,
|
| 584 |
+
"loss": 0.7839,
|
| 585 |
+
"step": 395
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.9197080291970803,
|
| 589 |
+
"grad_norm": 0.59765625,
|
| 590 |
+
"learning_rate": 8.829690815345886e-05,
|
| 591 |
+
"loss": 0.7781,
|
| 592 |
+
"step": 400
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.9562043795620436,
|
| 596 |
+
"grad_norm": 0.58203125,
|
| 597 |
+
"learning_rate": 8.57685161726715e-05,
|
| 598 |
+
"loss": 0.7457,
|
| 599 |
+
"step": 405
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.9927007299270074,
|
| 603 |
+
"grad_norm": 0.6171875,
|
| 604 |
+
"learning_rate": 8.324937766952638e-05,
|
| 605 |
+
"loss": 0.7623,
|
| 606 |
+
"step": 410
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 3.0,
|
| 610 |
+
"eval_loss": 2.275648355484009,
|
| 611 |
+
"eval_runtime": 0.9945,
|
| 612 |
+
"eval_samples_per_second": 5.028,
|
| 613 |
+
"eval_steps_per_second": 2.011,
|
| 614 |
+
"step": 411
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.0291970802919708,
|
| 618 |
+
"grad_norm": 0.8359375,
|
| 619 |
+
"learning_rate": 8.074113061741397e-05,
|
| 620 |
+
"loss": 0.7329,
|
| 621 |
+
"step": 415
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 3.065693430656934,
|
| 625 |
+
"grad_norm": 0.50390625,
|
| 626 |
+
"learning_rate": 7.824540590797568e-05,
|
| 627 |
+
"loss": 0.7052,
|
| 628 |
+
"step": 420
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 3.102189781021898,
|
| 632 |
+
"grad_norm": 0.5703125,
|
| 633 |
+
"learning_rate": 7.576382629067877e-05,
|
| 634 |
+
"loss": 0.7015,
|
| 635 |
+
"step": 425
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 3.1386861313868613,
|
| 639 |
+
"grad_norm": 0.6015625,
|
| 640 |
+
"learning_rate": 7.329800531768584e-05,
|
| 641 |
+
"loss": 0.696,
|
| 642 |
+
"step": 430
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 3.1751824817518246,
|
| 646 |
+
"grad_norm": 0.55078125,
|
| 647 |
+
"learning_rate": 7.084954629470417e-05,
|
| 648 |
+
"loss": 0.7154,
|
| 649 |
+
"step": 435
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 3.2116788321167884,
|
| 653 |
+
"grad_norm": 0.59765625,
|
| 654 |
+
"learning_rate": 6.842004123849752e-05,
|
| 655 |
+
"loss": 0.7113,
|
| 656 |
+
"step": 440
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 3.2481751824817517,
|
| 660 |
+
"grad_norm": 0.5625,
|
| 661 |
+
"learning_rate": 6.601106984173835e-05,
|
| 662 |
+
"loss": 0.7139,
|
| 663 |
+
"step": 445
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 3.2846715328467155,
|
| 667 |
+
"grad_norm": 0.59765625,
|
| 668 |
+
"learning_rate": 6.362419844587287e-05,
|
| 669 |
+
"loss": 0.6967,
|
| 670 |
+
"step": 450
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 3.321167883211679,
|
| 674 |
+
"grad_norm": 0.52734375,
|
| 675 |
+
"learning_rate": 6.126097902266772e-05,
|
| 676 |
+
"loss": 0.7073,
|
| 677 |
+
"step": 455
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 3.3576642335766422,
|
| 681 |
+
"grad_norm": 0.5625,
|
| 682 |
+
"learning_rate": 5.8922948165099524e-05,
|
| 683 |
+
"loss": 0.6857,
|
| 684 |
+
"step": 460
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.394160583941606,
|
| 688 |
+
"grad_norm": 0.55859375,
|
| 689 |
+
"learning_rate": 5.6611626088244194e-05,
|
| 690 |
+
"loss": 0.7199,
|
| 691 |
+
"step": 465
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 3.4306569343065694,
|
| 695 |
+
"grad_norm": 0.58203125,
|
| 696 |
+
"learning_rate": 5.432851564081534e-05,
|
| 697 |
+
"loss": 0.7075,
|
| 698 |
+
"step": 470
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 3.4671532846715327,
|
| 702 |
+
"grad_norm": 0.52734375,
|
| 703 |
+
"learning_rate": 5.207510132799436e-05,
|
| 704 |
+
"loss": 0.7006,
|
| 705 |
+
"step": 475
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.5036496350364965,
|
| 709 |
+
"grad_norm": 0.53515625,
|
| 710 |
+
"learning_rate": 4.9852848346187566e-05,
|
| 711 |
+
"loss": 0.7151,
|
| 712 |
+
"step": 480
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 3.54014598540146,
|
| 716 |
+
"grad_norm": 0.546875,
|
| 717 |
+
"learning_rate": 4.7663201630338816e-05,
|
| 718 |
+
"loss": 0.7129,
|
| 719 |
+
"step": 485
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 3.576642335766423,
|
| 723 |
+
"grad_norm": 0.5859375,
|
| 724 |
+
"learning_rate": 4.550758491441526e-05,
|
| 725 |
+
"loss": 0.7139,
|
| 726 |
+
"step": 490
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 3.613138686131387,
|
| 730 |
+
"grad_norm": 0.51953125,
|
| 731 |
+
"learning_rate": 4.3387399805679255e-05,
|
| 732 |
+
"loss": 0.7162,
|
| 733 |
+
"step": 495
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 3.6496350364963503,
|
| 737 |
+
"grad_norm": 0.55859375,
|
| 738 |
+
"learning_rate": 4.1304024873346705e-05,
|
| 739 |
+
"loss": 0.7132,
|
| 740 |
+
"step": 500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.686131386861314,
|
| 744 |
+
"grad_norm": 0.57421875,
|
| 745 |
+
"learning_rate": 3.9258814752225284e-05,
|
| 746 |
+
"loss": 0.7007,
|
| 747 |
+
"step": 505
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.7226277372262775,
|
| 751 |
+
"grad_norm": 0.546875,
|
| 752 |
+
"learning_rate": 3.725309926191479e-05,
|
| 753 |
+
"loss": 0.7037,
|
| 754 |
+
"step": 510
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.759124087591241,
|
| 758 |
+
"grad_norm": 0.73828125,
|
| 759 |
+
"learning_rate": 3.528818254214329e-05,
|
| 760 |
+
"loss": 0.7255,
|
| 761 |
+
"step": 515
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.795620437956204,
|
| 765 |
+
"grad_norm": 0.52734375,
|
| 766 |
+
"learning_rate": 3.336534220479961e-05,
|
| 767 |
+
"loss": 0.6966,
|
| 768 |
+
"step": 520
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.832116788321168,
|
| 772 |
+
"grad_norm": 0.5078125,
|
| 773 |
+
"learning_rate": 3.1485828503215585e-05,
|
| 774 |
+
"loss": 0.7143,
|
| 775 |
+
"step": 525
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.8686131386861313,
|
| 779 |
+
"grad_norm": 0.6328125,
|
| 780 |
+
"learning_rate": 2.9650863519236418e-05,
|
| 781 |
+
"loss": 0.7005,
|
| 782 |
+
"step": 530
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 3.905109489051095,
|
| 786 |
+
"grad_norm": 0.5703125,
|
| 787 |
+
"learning_rate": 2.7861640368608844e-05,
|
| 788 |
+
"loss": 0.7005,
|
| 789 |
+
"step": 535
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.9416058394160585,
|
| 793 |
+
"grad_norm": 0.53125,
|
| 794 |
+
"learning_rate": 2.6119322425203197e-05,
|
| 795 |
+
"loss": 0.7139,
|
| 796 |
+
"step": 540
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 3.978102189781022,
|
| 800 |
+
"grad_norm": 0.51953125,
|
| 801 |
+
"learning_rate": 2.4425042564574184e-05,
|
| 802 |
+
"loss": 0.709,
|
| 803 |
+
"step": 545
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 4.0,
|
| 807 |
+
"eval_loss": 2.341665267944336,
|
| 808 |
+
"eval_runtime": 0.9977,
|
| 809 |
+
"eval_samples_per_second": 5.012,
|
| 810 |
+
"eval_steps_per_second": 2.005,
|
| 811 |
+
"step": 548
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 4.014598540145985,
|
| 815 |
+
"grad_norm": 0.53515625,
|
| 816 |
+
"learning_rate": 2.277990242735185e-05,
|
| 817 |
+
"loss": 0.6801,
|
| 818 |
+
"step": 550
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 4.0510948905109485,
|
| 822 |
+
"grad_norm": 0.52734375,
|
| 823 |
+
"learning_rate": 2.118497170294195e-05,
|
| 824 |
+
"loss": 0.6495,
|
| 825 |
+
"step": 555
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 4.087591240875913,
|
| 829 |
+
"grad_norm": 0.5625,
|
| 830 |
+
"learning_rate": 1.9641287434001355e-05,
|
| 831 |
+
"loss": 0.672,
|
| 832 |
+
"step": 560
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 4.124087591240876,
|
| 836 |
+
"grad_norm": 0.55078125,
|
| 837 |
+
"learning_rate": 1.8149853342140645e-05,
|
| 838 |
+
"loss": 0.6611,
|
| 839 |
+
"step": 565
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.160583941605839,
|
| 843 |
+
"grad_norm": 0.59375,
|
| 844 |
+
"learning_rate": 1.671163917529285e-05,
|
| 845 |
+
"loss": 0.662,
|
| 846 |
+
"step": 570
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 4.197080291970803,
|
| 850 |
+
"grad_norm": 0.51171875,
|
| 851 |
+
"learning_rate": 1.5327580077171587e-05,
|
| 852 |
+
"loss": 0.6635,
|
| 853 |
+
"step": 575
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 4.233576642335766,
|
| 857 |
+
"grad_norm": 0.54296875,
|
| 858 |
+
"learning_rate": 1.3998575979229944e-05,
|
| 859 |
+
"loss": 0.6624,
|
| 860 |
+
"step": 580
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 4.2700729927007295,
|
| 864 |
+
"grad_norm": 0.50390625,
|
| 865 |
+
"learning_rate": 1.272549101551438e-05,
|
| 866 |
+
"loss": 0.6523,
|
| 867 |
+
"step": 585
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 4.306569343065694,
|
| 871 |
+
"grad_norm": 0.51171875,
|
| 872 |
+
"learning_rate": 1.1509152960794666e-05,
|
| 873 |
+
"loss": 0.6607,
|
| 874 |
+
"step": 590
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.343065693430657,
|
| 878 |
+
"grad_norm": 0.546875,
|
| 879 |
+
"learning_rate": 1.035035269233493e-05,
|
| 880 |
+
"loss": 0.6626,
|
| 881 |
+
"step": 595
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.37956204379562,
|
| 885 |
+
"grad_norm": 0.54296875,
|
| 886 |
+
"learning_rate": 9.249843675656212e-06,
|
| 887 |
+
"loss": 0.678,
|
| 888 |
+
"step": 600
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 4.416058394160584,
|
| 892 |
+
"grad_norm": 0.5234375,
|
| 893 |
+
"learning_rate": 8.208341474624071e-06,
|
| 894 |
+
"loss": 0.6783,
|
| 895 |
+
"step": 605
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 4.452554744525547,
|
| 899 |
+
"grad_norm": 0.53515625,
|
| 900 |
+
"learning_rate": 7.226523286180776e-06,
|
| 901 |
+
"loss": 0.6699,
|
| 902 |
+
"step": 610
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 4.489051094890511,
|
| 906 |
+
"grad_norm": 0.5703125,
|
| 907 |
+
"learning_rate": 6.3050275000238414e-06,
|
| 908 |
+
"loss": 0.6607,
|
| 909 |
+
"step": 615
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 4.525547445255475,
|
| 913 |
+
"grad_norm": 0.5234375,
|
| 914 |
+
"learning_rate": 5.4444532835175144e-06,
|
| 915 |
+
"loss": 0.6702,
|
| 916 |
+
"step": 620
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 4.562043795620438,
|
| 920 |
+
"grad_norm": 0.5234375,
|
| 921 |
+
"learning_rate": 4.6453601921072395e-06,
|
| 922 |
+
"loss": 0.6793,
|
| 923 |
+
"step": 625
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 4.598540145985401,
|
| 927 |
+
"grad_norm": 0.5234375,
|
| 928 |
+
"learning_rate": 3.908267805490051e-06,
|
| 929 |
+
"loss": 0.6622,
|
| 930 |
+
"step": 630
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 4.635036496350365,
|
| 934 |
+
"grad_norm": 0.54296875,
|
| 935 |
+
"learning_rate": 3.233655389777801e-06,
|
| 936 |
+
"loss": 0.677,
|
| 937 |
+
"step": 635
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 4.671532846715328,
|
| 941 |
+
"grad_norm": 0.5234375,
|
| 942 |
+
"learning_rate": 2.62196158587269e-06,
|
| 943 |
+
"loss": 0.6588,
|
| 944 |
+
"step": 640
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 4.708029197080292,
|
| 948 |
+
"grad_norm": 0.5234375,
|
| 949 |
+
"learning_rate": 2.073584124257899e-06,
|
| 950 |
+
"loss": 0.6621,
|
| 951 |
+
"step": 645
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 4.744525547445256,
|
| 955 |
+
"grad_norm": 0.53515625,
|
| 956 |
+
"learning_rate": 1.5888795663883904e-06,
|
| 957 |
+
"loss": 0.6655,
|
| 958 |
+
"step": 650
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 4.781021897810219,
|
| 962 |
+
"grad_norm": 0.515625,
|
| 963 |
+
"learning_rate": 1.1681630728506699e-06,
|
| 964 |
+
"loss": 0.6653,
|
| 965 |
+
"step": 655
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 4.817518248175182,
|
| 969 |
+
"grad_norm": 0.52734375,
|
| 970 |
+
"learning_rate": 8.117081984415298e-07,
|
| 971 |
+
"loss": 0.6734,
|
| 972 |
+
"step": 660
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 4.854014598540146,
|
| 976 |
+
"grad_norm": 0.5390625,
|
| 977 |
+
"learning_rate": 5.19746714299596e-07,
|
| 978 |
+
"loss": 0.6541,
|
| 979 |
+
"step": 665
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 4.89051094890511,
|
| 983 |
+
"grad_norm": 0.5390625,
|
| 984 |
+
"learning_rate": 2.9246845720496407e-07,
|
| 985 |
+
"loss": 0.6722,
|
| 986 |
+
"step": 670
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 4.927007299270073,
|
| 990 |
+
"grad_norm": 0.55859375,
|
| 991 |
+
"learning_rate": 1.300212061451367e-07,
|
| 992 |
+
"loss": 0.6472,
|
| 993 |
+
"step": 675
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 4.963503649635037,
|
| 997 |
+
"grad_norm": 0.51953125,
|
| 998 |
+
"learning_rate": 3.251058622737446e-08,
|
| 999 |
+
"loss": 0.667,
|
| 1000 |
+
"step": 680
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 5.0,
|
| 1004 |
+
"grad_norm": 0.52734375,
|
| 1005 |
+
"learning_rate": 0.0,
|
| 1006 |
+
"loss": 0.6601,
|
| 1007 |
+
"step": 685
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 5.0,
|
| 1011 |
+
"eval_loss": 2.3811252117156982,
|
| 1012 |
+
"eval_runtime": 0.9953,
|
| 1013 |
+
"eval_samples_per_second": 5.024,
|
| 1014 |
+
"eval_steps_per_second": 2.01,
|
| 1015 |
+
"step": 685
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 5.0,
|
| 1019 |
+
"step": 685,
|
| 1020 |
+
"total_flos": 1.0472781231601746e+18,
|
| 1021 |
+
"train_loss": 2.151051264783762,
|
| 1022 |
+
"train_runtime": 5341.9856,
|
| 1023 |
+
"train_samples_per_second": 2.052,
|
| 1024 |
+
"train_steps_per_second": 0.128
|
| 1025 |
}
|
| 1026 |
],
|
| 1027 |
"logging_steps": 5,
|
| 1028 |
+
"max_steps": 685,
|
| 1029 |
"num_input_tokens_seen": 0,
|
| 1030 |
+
"num_train_epochs": 5,
|
| 1031 |
"save_steps": 100,
|
| 1032 |
"stateful_callbacks": {
|
| 1033 |
"TrainerControl": {
|
|
|
|
| 1041 |
"attributes": {}
|
| 1042 |
}
|
| 1043 |
},
|
| 1044 |
+
"total_flos": 1.0472781231601746e+18,
|
| 1045 |
"train_batch_size": 4,
|
| 1046 |
"trial_name": null,
|
| 1047 |
"trial_params": null
|