Commit
•
a30ad28
1
Parent(s):
8d227f6
Model save
Browse files- README.md +14 -18
- all_results.json +6 -11
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- train_results.json +6 -6
- trainer_state.json +901 -110
README.md
CHANGED
|
@@ -2,10 +2,6 @@
|
|
| 2 |
license: gemma
|
| 3 |
base_model: google/gemma-2-2b
|
| 4 |
tags:
|
| 5 |
-
- easylm
|
| 6 |
-
- trl
|
| 7 |
-
- sft
|
| 8 |
-
- generated_from_trainer
|
| 9 |
- trl
|
| 10 |
- sft
|
| 11 |
- generated_from_trainer
|
|
@@ -23,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 23 |
|
| 24 |
This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on the alpaca_farm dataset.
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
-
- Loss: 0.
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
@@ -52,24 +48,24 @@ The following hyperparameters were used during training:
|
|
| 52 |
- total_eval_batch_size: 16
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: cosine
|
| 55 |
-
- num_epochs:
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
base_model: google/gemma-2-2b
|
| 4 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
|
|
|
| 19 |
|
| 20 |
This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on the alpaca_farm dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.7070
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 48 |
- total_eval_batch_size: 16
|
| 49 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
- lr_scheduler_type: cosine
|
| 51 |
+
- num_epochs: 2
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
+
| 0.6727 | 0.16 | 100 | 0.6869 |
|
| 58 |
+
| 0.6056 | 0.32 | 200 | 0.6831 |
|
| 59 |
+
| 0.7033 | 0.48 | 300 | 0.6797 |
|
| 60 |
+
| 0.6786 | 0.64 | 400 | 0.6771 |
|
| 61 |
+
| 0.6476 | 0.8 | 500 | 0.6736 |
|
| 62 |
+
| 0.6562 | 0.96 | 600 | 0.6708 |
|
| 63 |
+
| 0.461 | 1.12 | 700 | 0.7041 |
|
| 64 |
+
| 0.4578 | 1.28 | 800 | 0.7093 |
|
| 65 |
+
| 0.4817 | 1.44 | 900 | 0.7055 |
|
| 66 |
+
| 0.4324 | 1.6 | 1000 | 0.7080 |
|
| 67 |
+
| 0.4693 | 1.76 | 1100 | 0.7081 |
|
| 68 |
+
| 0.4475 | 1.92 | 1200 | 0.7070 |
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 277.295,
|
| 7 |
-
"eval_steps_per_second": 17.331,
|
| 8 |
-
"total_flos": 2.283996277689549e+16,
|
| 9 |
-
"train_loss": 0.6787182281494141,
|
| 10 |
-
"train_runtime": 858.483,
|
| 11 |
"train_samples": 10000,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 4.566411071245517e+16,
|
| 4 |
+
"train_loss": 0.5696872653961181,
|
| 5 |
+
"train_runtime": 1519.873,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 13.159,
|
| 8 |
+
"train_steps_per_second": 0.822
|
| 9 |
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4988025760
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:661307a70cfa0d3186eb20155bbd950fd35297d5138a4afdcbd550e94c9eb752
|
| 3 |
size 4988025760
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1420339880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d28227d7bf815ed9760efe3c3a0375227a9a0d7be7e189a0ea9a0dee8214db1
|
| 3 |
size 1420339880
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 10000,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 4.566411071245517e+16,
|
| 4 |
+
"train_loss": 0.5696872653961181,
|
| 5 |
+
"train_runtime": 1519.873,
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 13.159,
|
| 8 |
+
"train_steps_per_second": 0.822
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,207 +1,998 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 2.
|
| 15 |
-
"loss":
|
| 16 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.08,
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
| 23 |
-
"eval_steps_per_second": 13.658,
|
| 24 |
"step": 50
|
| 25 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
{
|
| 27 |
"epoch": 0.16,
|
| 28 |
-
"grad_norm": 4.
|
| 29 |
-
"learning_rate": 2.
|
| 30 |
-
"loss": 0.
|
| 31 |
"step": 100
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.16,
|
| 35 |
-
"eval_loss": 0.
|
| 36 |
-
"eval_runtime":
|
| 37 |
-
"eval_samples_per_second":
|
| 38 |
-
"eval_steps_per_second":
|
| 39 |
"step": 100
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate": 2.
|
| 45 |
-
"loss": 0.
|
| 46 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 0.24,
|
| 50 |
-
"
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"eval_steps_per_second": 16.974,
|
| 54 |
"step": 150
|
| 55 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
{
|
| 57 |
"epoch": 0.32,
|
| 58 |
-
"grad_norm": 4.
|
| 59 |
-
"learning_rate": 2.
|
| 60 |
-
"loss": 0.
|
| 61 |
"step": 200
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 0.32,
|
| 65 |
-
"eval_loss": 0.
|
| 66 |
-
"eval_runtime":
|
| 67 |
-
"eval_samples_per_second":
|
| 68 |
-
"eval_steps_per_second":
|
| 69 |
"step": 200
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"grad_norm": 4.
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"loss": 0.
|
| 76 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 0.4,
|
| 80 |
-
"
|
| 81 |
-
"
|
| 82 |
-
"
|
| 83 |
-
"eval_steps_per_second": 17.232,
|
| 84 |
"step": 250
|
| 85 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
{
|
| 87 |
"epoch": 0.48,
|
| 88 |
-
"grad_norm": 4.
|
| 89 |
-
"learning_rate":
|
| 90 |
-
"loss": 0.
|
| 91 |
"step": 300
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 0.48,
|
| 95 |
-
"eval_loss": 0.
|
| 96 |
-
"eval_runtime":
|
| 97 |
-
"eval_samples_per_second":
|
| 98 |
-
"eval_steps_per_second":
|
| 99 |
"step": 300
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
-
"grad_norm": 4.
|
| 104 |
-
"learning_rate":
|
| 105 |
-
"loss": 0.
|
| 106 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
},
|
| 108 |
{
|
| 109 |
"epoch": 0.56,
|
| 110 |
-
"
|
| 111 |
-
"
|
| 112 |
-
"
|
| 113 |
-
"eval_steps_per_second": 15.981,
|
| 114 |
"step": 350
|
| 115 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
{
|
| 117 |
"epoch": 0.64,
|
| 118 |
-
"grad_norm": 4.
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 400
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.64,
|
| 125 |
-
"eval_loss": 0.
|
| 126 |
-
"eval_runtime":
|
| 127 |
-
"eval_samples_per_second":
|
| 128 |
-
"eval_steps_per_second":
|
| 129 |
"step": 400
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm": 4.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 0.
|
| 136 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.72,
|
| 140 |
-
"
|
| 141 |
-
"
|
| 142 |
-
"
|
| 143 |
-
"eval_steps_per_second": 17.41,
|
| 144 |
"step": 450
|
| 145 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
{
|
| 147 |
"epoch": 0.8,
|
| 148 |
-
"grad_norm": 4.
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 500
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.8,
|
| 155 |
-
"eval_loss": 0.
|
| 156 |
-
"eval_runtime": 7.
|
| 157 |
-
"eval_samples_per_second":
|
| 158 |
-
"eval_steps_per_second": 17.
|
| 159 |
"step": 500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
-
"grad_norm": 4.
|
| 164 |
-
"learning_rate": 1.
|
| 165 |
-
"loss": 0.
|
| 166 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 0.88,
|
| 170 |
-
"
|
| 171 |
-
"
|
| 172 |
-
"
|
| 173 |
-
"eval_steps_per_second": 17.037,
|
| 174 |
"step": 550
|
| 175 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
{
|
| 177 |
"epoch": 0.96,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 1.
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 600
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.96,
|
| 185 |
-
"eval_loss": 0.
|
| 186 |
-
"eval_runtime": 7.
|
| 187 |
-
"eval_samples_per_second":
|
| 188 |
-
"eval_steps_per_second": 16.
|
| 189 |
"step": 600
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"
|
| 194 |
-
"
|
| 195 |
-
"
|
| 196 |
-
"
|
| 197 |
-
|
| 198 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 199 |
}
|
| 200 |
],
|
| 201 |
-
"logging_steps":
|
| 202 |
-
"max_steps":
|
| 203 |
"num_input_tokens_seen": 0,
|
| 204 |
-
"num_train_epochs":
|
| 205 |
"save_steps": 500,
|
| 206 |
"stateful_callbacks": {
|
| 207 |
"TrainerControl": {
|
|
@@ -215,7 +1006,7 @@
|
|
| 215 |
"attributes": {}
|
| 216 |
}
|
| 217 |
},
|
| 218 |
-
"total_flos":
|
| 219 |
"train_batch_size": 2,
|
| 220 |
"trial_name": null,
|
| 221 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1250,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.016,
|
| 13 |
+
"grad_norm": 6.320438861846924,
|
| 14 |
+
"learning_rate": 2.99952628392495e-06,
|
| 15 |
+
"loss": 1.3054,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.032,
|
| 20 |
+
"grad_norm": 5.475564956665039,
|
| 21 |
+
"learning_rate": 2.9981054349090266e-06,
|
| 22 |
+
"loss": 0.6821,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.048,
|
| 27 |
+
"grad_norm": 5.251683235168457,
|
| 28 |
+
"learning_rate": 2.995738350390921e-06,
|
| 29 |
+
"loss": 0.6935,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.064,
|
| 34 |
+
"grad_norm": 5.057413578033447,
|
| 35 |
+
"learning_rate": 2.9924265254719506e-06,
|
| 36 |
+
"loss": 0.6776,
|
| 37 |
+
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.08,
|
| 41 |
+
"grad_norm": 4.680449962615967,
|
| 42 |
+
"learning_rate": 2.988172051971717e-06,
|
| 43 |
+
"loss": 0.6582,
|
|
|
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.096,
|
| 48 |
+
"grad_norm": 5.521228313446045,
|
| 49 |
+
"learning_rate": 2.982977617106871e-06,
|
| 50 |
+
"loss": 0.6972,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.112,
|
| 55 |
+
"grad_norm": 4.839247226715088,
|
| 56 |
+
"learning_rate": 2.9768465017938084e-06,
|
| 57 |
+
"loss": 0.6749,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.128,
|
| 62 |
+
"grad_norm": 5.12637996673584,
|
| 63 |
+
"learning_rate": 2.9697825785763704e-06,
|
| 64 |
+
"loss": 0.6863,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.144,
|
| 69 |
+
"grad_norm": 4.7851152420043945,
|
| 70 |
+
"learning_rate": 2.961790309179866e-06,
|
| 71 |
+
"loss": 0.6785,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
{
|
| 75 |
"epoch": 0.16,
|
| 76 |
+
"grad_norm": 4.761219501495361,
|
| 77 |
+
"learning_rate": 2.9528747416929465e-06,
|
| 78 |
+
"loss": 0.6727,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.16,
|
| 83 |
+
"eval_loss": 0.6868906021118164,
|
| 84 |
+
"eval_runtime": 8.6852,
|
| 85 |
+
"eval_samples_per_second": 230.276,
|
| 86 |
+
"eval_steps_per_second": 14.392,
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.176,
|
| 91 |
+
"grad_norm": 4.815207004547119,
|
| 92 |
+
"learning_rate": 2.943041507379129e-06,
|
| 93 |
+
"loss": 0.6787,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.192,
|
| 98 |
+
"grad_norm": 4.663937568664551,
|
| 99 |
+
"learning_rate": 2.9322968171199645e-06,
|
| 100 |
+
"loss": 0.6772,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.208,
|
| 105 |
+
"grad_norm": 5.208789348602295,
|
| 106 |
+
"learning_rate": 2.9206474574921165e-06,
|
| 107 |
+
"loss": 0.656,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.224,
|
| 112 |
+
"grad_norm": 4.607890605926514,
|
| 113 |
+
"learning_rate": 2.9081007864808113e-06,
|
| 114 |
+
"loss": 0.6792,
|
| 115 |
+
"step": 140
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.24,
|
| 119 |
+
"grad_norm": 5.205310821533203,
|
| 120 |
+
"learning_rate": 2.894664728832377e-06,
|
| 121 |
+
"loss": 0.6989,
|
|
|
|
| 122 |
"step": 150
|
| 123 |
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.256,
|
| 126 |
+
"grad_norm": 4.700341701507568,
|
| 127 |
+
"learning_rate": 2.8803477710488056e-06,
|
| 128 |
+
"loss": 0.6673,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.272,
|
| 133 |
+
"grad_norm": 4.271303653717041,
|
| 134 |
+
"learning_rate": 2.8651589560274937e-06,
|
| 135 |
+
"loss": 0.6743,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.288,
|
| 140 |
+
"grad_norm": 4.5935468673706055,
|
| 141 |
+
"learning_rate": 2.8491078773495566e-06,
|
| 142 |
+
"loss": 0.6634,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.304,
|
| 147 |
+
"grad_norm": 4.505198955535889,
|
| 148 |
+
"learning_rate": 2.832204673220317e-06,
|
| 149 |
+
"loss": 0.6487,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
{
|
| 153 |
"epoch": 0.32,
|
| 154 |
+
"grad_norm": 4.946630477905273,
|
| 155 |
+
"learning_rate": 2.814460020065795e-06,
|
| 156 |
+
"loss": 0.6056,
|
| 157 |
"step": 200
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 0.32,
|
| 161 |
+
"eval_loss": 0.6831114292144775,
|
| 162 |
+
"eval_runtime": 9.3989,
|
| 163 |
+
"eval_samples_per_second": 212.791,
|
| 164 |
+
"eval_steps_per_second": 13.299,
|
| 165 |
"step": 200
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 0.336,
|
| 169 |
+
"grad_norm": 4.913090229034424,
|
| 170 |
+
"learning_rate": 2.795885125789253e-06,
|
| 171 |
+
"loss": 0.652,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.352,
|
| 176 |
+
"grad_norm": 4.906637191772461,
|
| 177 |
+
"learning_rate": 2.776491722692038e-06,
|
| 178 |
+
"loss": 0.6599,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.368,
|
| 183 |
+
"grad_norm": 4.214770317077637,
|
| 184 |
+
"learning_rate": 2.756292060063213e-06,
|
| 185 |
+
"loss": 0.6539,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.384,
|
| 190 |
+
"grad_norm": 4.808592796325684,
|
| 191 |
+
"learning_rate": 2.735298896442641e-06,
|
| 192 |
+
"loss": 0.6666,
|
| 193 |
+
"step": 240
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 0.4,
|
| 197 |
+
"grad_norm": 4.6029839515686035,
|
| 198 |
+
"learning_rate": 2.713525491562421e-06,
|
| 199 |
+
"loss": 0.6924,
|
|
|
|
| 200 |
"step": 250
|
| 201 |
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.416,
|
| 204 |
+
"grad_norm": 4.828456878662109,
|
| 205 |
+
"learning_rate": 2.690985597971753e-06,
|
| 206 |
+
"loss": 0.6937,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.432,
|
| 211 |
+
"grad_norm": 4.152997970581055,
|
| 212 |
+
"learning_rate": 2.6676934523505355e-06,
|
| 213 |
+
"loss": 0.6664,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.448,
|
| 218 |
+
"grad_norm": 4.614770412445068,
|
| 219 |
+
"learning_rate": 2.643663766517172e-06,
|
| 220 |
+
"loss": 0.6722,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.464,
|
| 225 |
+
"grad_norm": 4.493509292602539,
|
| 226 |
+
"learning_rate": 2.6189117181362736e-06,
|
| 227 |
+
"loss": 0.6689,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
{
|
| 231 |
"epoch": 0.48,
|
| 232 |
+
"grad_norm": 4.5999274253845215,
|
| 233 |
+
"learning_rate": 2.5934529411321173e-06,
|
| 234 |
+
"loss": 0.7033,
|
| 235 |
"step": 300
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 0.48,
|
| 239 |
+
"eval_loss": 0.6796725988388062,
|
| 240 |
+
"eval_runtime": 9.6974,
|
| 241 |
+
"eval_samples_per_second": 206.241,
|
| 242 |
+
"eval_steps_per_second": 12.89,
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 0.496,
|
| 247 |
+
"grad_norm": 4.2827935218811035,
|
| 248 |
+
"learning_rate": 2.5673035158139285e-06,
|
| 249 |
+
"loss": 0.6596,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.512,
|
| 254 |
+
"grad_norm": 4.7542805671691895,
|
| 255 |
+
"learning_rate": 2.5404799587192076e-06,
|
| 256 |
+
"loss": 0.6549,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.528,
|
| 261 |
+
"grad_norm": 4.7150492668151855,
|
| 262 |
+
"learning_rate": 2.5129992121815365e-06,
|
| 263 |
+
"loss": 0.6757,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.544,
|
| 268 |
+
"grad_norm": 4.777836322784424,
|
| 269 |
+
"learning_rate": 2.484878633629435e-06,
|
| 270 |
+
"loss": 0.6806,
|
| 271 |
+
"step": 340
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 0.56,
|
| 275 |
+
"grad_norm": 4.323108196258545,
|
| 276 |
+
"learning_rate": 2.456135984623035e-06,
|
| 277 |
+
"loss": 0.6629,
|
|
|
|
| 278 |
"step": 350
|
| 279 |
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.576,
|
| 282 |
+
"grad_norm": 4.968432426452637,
|
| 283 |
+
"learning_rate": 2.4267894196355018e-06,
|
| 284 |
+
"loss": 0.6858,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.592,
|
| 289 |
+
"grad_norm": 4.666640758514404,
|
| 290 |
+
"learning_rate": 2.3968574745862785e-06,
|
| 291 |
+
"loss": 0.665,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.608,
|
| 296 |
+
"grad_norm": 4.655545234680176,
|
| 297 |
+
"learning_rate": 2.3663590551334015e-06,
|
| 298 |
+
"loss": 0.6661,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.624,
|
| 303 |
+
"grad_norm": 4.554004669189453,
|
| 304 |
+
"learning_rate": 2.3353134247322823e-06,
|
| 305 |
+
"loss": 0.6559,
|
| 306 |
+
"step": 390
|
| 307 |
+
},
|
| 308 |
{
|
| 309 |
"epoch": 0.64,
|
| 310 |
+
"grad_norm": 4.362437725067139,
|
| 311 |
+
"learning_rate": 2.303740192468495e-06,
|
| 312 |
+
"loss": 0.6786,
|
| 313 |
"step": 400
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 0.64,
|
| 317 |
+
"eval_loss": 0.6770616173744202,
|
| 318 |
+
"eval_runtime": 7.3123,
|
| 319 |
+
"eval_samples_per_second": 273.512,
|
| 320 |
+
"eval_steps_per_second": 17.095,
|
| 321 |
"step": 400
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 0.656,
|
| 325 |
+
"grad_norm": 4.237730503082275,
|
| 326 |
+
"learning_rate": 2.2716593006722595e-06,
|
| 327 |
+
"loss": 0.6456,
|
| 328 |
+
"step": 410
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.672,
|
| 332 |
+
"grad_norm": 4.062387943267822,
|
| 333 |
+
"learning_rate": 2.2390910123224374e-06,
|
| 334 |
+
"loss": 0.6767,
|
| 335 |
+
"step": 420
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.688,
|
| 339 |
+
"grad_norm": 4.175485134124756,
|
| 340 |
+
"learning_rate": 2.2060558982479992e-06,
|
| 341 |
+
"loss": 0.6389,
|
| 342 |
+
"step": 430
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.704,
|
| 346 |
+
"grad_norm": 4.427236080169678,
|
| 347 |
+
"learning_rate": 2.1725748241350487e-06,
|
| 348 |
+
"loss": 0.6403,
|
| 349 |
+
"step": 440
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 0.72,
|
| 353 |
+
"grad_norm": 4.3691511154174805,
|
| 354 |
+
"learning_rate": 2.138668937347609e-06,
|
| 355 |
+
"loss": 0.6559,
|
|
|
|
| 356 |
"step": 450
|
| 357 |
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.736,
|
| 360 |
+
"grad_norm": 4.3593430519104,
|
| 361 |
+
"learning_rate": 2.1043596535704943e-06,
|
| 362 |
+
"loss": 0.6787,
|
| 363 |
+
"step": 460
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.752,
|
| 367 |
+
"grad_norm": 4.350930213928223,
|
| 368 |
+
"learning_rate": 2.069668643282702e-06,
|
| 369 |
+
"loss": 0.6975,
|
| 370 |
+
"step": 470
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.768,
|
| 374 |
+
"grad_norm": 4.215636730194092,
|
| 375 |
+
"learning_rate": 2.034617818069876e-06,
|
| 376 |
+
"loss": 0.684,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.784,
|
| 381 |
+
"grad_norm": 3.999462127685547,
|
| 382 |
+
"learning_rate": 1.99922931678448e-06,
|
| 383 |
+
"loss": 0.6607,
|
| 384 |
+
"step": 490
|
| 385 |
+
},
|
| 386 |
{
|
| 387 |
"epoch": 0.8,
|
| 388 |
+
"grad_norm": 4.186202526092529,
|
| 389 |
+
"learning_rate": 1.963525491562421e-06,
|
| 390 |
+
"loss": 0.6476,
|
| 391 |
"step": 500
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 0.8,
|
| 395 |
+
"eval_loss": 0.6736403107643127,
|
| 396 |
+
"eval_runtime": 7.3017,
|
| 397 |
+
"eval_samples_per_second": 273.908,
|
| 398 |
+
"eval_steps_per_second": 17.119,
|
| 399 |
"step": 500
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 0.816,
|
| 403 |
+
"grad_norm": 4.060866355895996,
|
| 404 |
+
"learning_rate": 1.927528893704964e-06,
|
| 405 |
+
"loss": 0.6591,
|
| 406 |
+
"step": 510
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.832,
|
| 410 |
+
"grad_norm": 4.125776767730713,
|
| 411 |
+
"learning_rate": 1.8912622594348455e-06,
|
| 412 |
+
"loss": 0.6922,
|
| 413 |
+
"step": 520
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.848,
|
| 417 |
+
"grad_norm": 3.8623437881469727,
|
| 418 |
+
"learning_rate": 1.8547484955355872e-06,
|
| 419 |
+
"loss": 0.6513,
|
| 420 |
+
"step": 530
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.864,
|
| 424 |
+
"grad_norm": 4.550040245056152,
|
| 425 |
+
"learning_rate": 1.8180106648830824e-06,
|
| 426 |
+
"loss": 0.663,
|
| 427 |
+
"step": 540
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 0.88,
|
| 431 |
+
"grad_norm": 4.267393589019775,
|
| 432 |
+
"learning_rate": 1.7810719718785873e-06,
|
| 433 |
+
"loss": 0.6685,
|
|
|
|
| 434 |
"step": 550
|
| 435 |
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.896,
|
| 438 |
+
"grad_norm": 4.428436756134033,
|
| 439 |
+
"learning_rate": 1.7439557477923257e-06,
|
| 440 |
+
"loss": 0.7051,
|
| 441 |
+
"step": 560
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.912,
|
| 445 |
+
"grad_norm": 4.695955276489258,
|
| 446 |
+
"learning_rate": 1.706685436026957e-06,
|
| 447 |
+
"loss": 0.6251,
|
| 448 |
+
"step": 570
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.928,
|
| 452 |
+
"grad_norm": 4.578221797943115,
|
| 453 |
+
"learning_rate": 1.6692845773102223e-06,
|
| 454 |
+
"loss": 0.6569,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.944,
|
| 459 |
+
"grad_norm": 4.15753173828125,
|
| 460 |
+
"learning_rate": 1.6317767948261151e-06,
|
| 461 |
+
"loss": 0.6451,
|
| 462 |
+
"step": 590
|
| 463 |
+
},
|
| 464 |
{
|
| 465 |
"epoch": 0.96,
|
| 466 |
+
"grad_norm": 3.946852922439575,
|
| 467 |
+
"learning_rate": 1.5941857792939703e-06,
|
| 468 |
+
"loss": 0.6562,
|
| 469 |
"step": 600
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 0.96,
|
| 473 |
+
"eval_loss": 0.6707749962806702,
|
| 474 |
+
"eval_runtime": 7.4417,
|
| 475 |
+
"eval_samples_per_second": 268.756,
|
| 476 |
+
"eval_steps_per_second": 16.797,
|
| 477 |
"step": 600
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 0.976,
|
| 481 |
+
"grad_norm": 4.214157581329346,
|
| 482 |
+
"learning_rate": 1.556535274004902e-06,
|
| 483 |
+
"loss": 0.6556,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.992,
|
| 488 |
+
"grad_norm": 4.3779616355896,
|
| 489 |
+
"learning_rate": 1.518849059825029e-06,
|
| 490 |
+
"loss": 0.6764,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.008,
|
| 495 |
+
"grad_norm": 3.7831478118896484,
|
| 496 |
+
"learning_rate": 1.481150940174971e-06,
|
| 497 |
+
"loss": 0.5361,
|
| 498 |
+
"step": 630
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.024,
|
| 502 |
+
"grad_norm": 4.746358394622803,
|
| 503 |
+
"learning_rate": 1.4434647259950982e-06,
|
| 504 |
+
"loss": 0.4773,
|
| 505 |
+
"step": 640
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.04,
|
| 509 |
+
"grad_norm": 4.30097770690918,
|
| 510 |
+
"learning_rate": 1.40581422070603e-06,
|
| 511 |
+
"loss": 0.4829,
|
| 512 |
+
"step": 650
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.056,
|
| 516 |
+
"grad_norm": 4.781858444213867,
|
| 517 |
+
"learning_rate": 1.3682232051738854e-06,
|
| 518 |
+
"loss": 0.4635,
|
| 519 |
+
"step": 660
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.072,
|
| 523 |
+
"grad_norm": 4.712700366973877,
|
| 524 |
+
"learning_rate": 1.3307154226897775e-06,
|
| 525 |
+
"loss": 0.4775,
|
| 526 |
+
"step": 670
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.088,
|
| 530 |
+
"grad_norm": 4.640942573547363,
|
| 531 |
+
"learning_rate": 1.293314563973043e-06,
|
| 532 |
+
"loss": 0.4421,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.104,
|
| 537 |
+
"grad_norm": 5.104272365570068,
|
| 538 |
+
"learning_rate": 1.2560442522076746e-06,
|
| 539 |
+
"loss": 0.446,
|
| 540 |
+
"step": 690
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.12,
|
| 544 |
+
"grad_norm": 5.069291591644287,
|
| 545 |
+
"learning_rate": 1.2189280281214128e-06,
|
| 546 |
+
"loss": 0.461,
|
| 547 |
+
"step": 700
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.12,
|
| 551 |
+
"eval_loss": 0.7041329145431519,
|
| 552 |
+
"eval_runtime": 7.9231,
|
| 553 |
+
"eval_samples_per_second": 252.428,
|
| 554 |
+
"eval_steps_per_second": 15.777,
|
| 555 |
+
"step": 700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.1360000000000001,
|
| 559 |
+
"grad_norm": 5.07060432434082,
|
| 560 |
+
"learning_rate": 1.1819893351169183e-06,
|
| 561 |
+
"loss": 0.4614,
|
| 562 |
+
"step": 710
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.152,
|
| 566 |
+
"grad_norm": 6.156511306762695,
|
| 567 |
+
"learning_rate": 1.1452515044644133e-06,
|
| 568 |
+
"loss": 0.4785,
|
| 569 |
+
"step": 720
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.168,
|
| 573 |
+
"grad_norm": 5.145558834075928,
|
| 574 |
+
"learning_rate": 1.108737740565155e-06,
|
| 575 |
+
"loss": 0.4463,
|
| 576 |
+
"step": 730
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.184,
|
| 580 |
+
"grad_norm": 4.276706218719482,
|
| 581 |
+
"learning_rate": 1.0724711062950359e-06,
|
| 582 |
+
"loss": 0.4595,
|
| 583 |
+
"step": 740
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.2,
|
| 587 |
+
"grad_norm": 5.321038722991943,
|
| 588 |
+
"learning_rate": 1.036474508437579e-06,
|
| 589 |
+
"loss": 0.4687,
|
| 590 |
+
"step": 750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.216,
|
| 594 |
+
"grad_norm": 4.610326766967773,
|
| 595 |
+
"learning_rate": 1.0007706832155202e-06,
|
| 596 |
+
"loss": 0.4293,
|
| 597 |
+
"step": 760
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.232,
|
| 601 |
+
"grad_norm": 5.0618462562561035,
|
| 602 |
+
"learning_rate": 9.65382181930124e-07,
|
| 603 |
+
"loss": 0.4636,
|
| 604 |
+
"step": 770
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.248,
|
| 608 |
+
"grad_norm": 4.6133036613464355,
|
| 609 |
+
"learning_rate": 9.303313567172986e-07,
|
| 610 |
+
"loss": 0.4578,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.264,
|
| 615 |
+
"grad_norm": 4.833690166473389,
|
| 616 |
+
"learning_rate": 8.956403464295061e-07,
|
| 617 |
+
"loss": 0.4511,
|
| 618 |
+
"step": 790
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.28,
|
| 622 |
+
"grad_norm": 4.997283935546875,
|
| 623 |
+
"learning_rate": 8.613310626523911e-07,
|
| 624 |
+
"loss": 0.4578,
|
| 625 |
+
"step": 800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.28,
|
| 629 |
+
"eval_loss": 0.7093445062637329,
|
| 630 |
+
"eval_runtime": 9.5227,
|
| 631 |
+
"eval_samples_per_second": 210.025,
|
| 632 |
+
"eval_steps_per_second": 13.127,
|
| 633 |
+
"step": 800
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.296,
|
| 637 |
+
"grad_norm": 5.087992191314697,
|
| 638 |
+
"learning_rate": 8.274251758649519e-07,
|
| 639 |
+
"loss": 0.4858,
|
| 640 |
+
"step": 810
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.312,
|
| 644 |
+
"grad_norm": 5.305168628692627,
|
| 645 |
+
"learning_rate": 7.939441017520012e-07,
|
| 646 |
+
"loss": 0.4756,
|
| 647 |
+
"step": 820
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.328,
|
| 651 |
+
"grad_norm": 4.691494464874268,
|
| 652 |
+
"learning_rate": 7.609089876775628e-07,
|
| 653 |
+
"loss": 0.4593,
|
| 654 |
+
"step": 830
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.3439999999999999,
|
| 658 |
+
"grad_norm": 6.369288921356201,
|
| 659 |
+
"learning_rate": 7.283406993277403e-07,
|
| 660 |
+
"loss": 0.4461,
|
| 661 |
+
"step": 840
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.3599999999999999,
|
| 665 |
+
"grad_norm": 4.588751792907715,
|
| 666 |
+
"learning_rate": 6.962598075315047e-07,
|
| 667 |
+
"loss": 0.4569,
|
| 668 |
+
"step": 850
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.376,
|
| 672 |
+
"grad_norm": 4.815459728240967,
|
| 673 |
+
"learning_rate": 6.646865752677186e-07,
|
| 674 |
+
"loss": 0.4209,
|
| 675 |
+
"step": 860
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.392,
|
| 679 |
+
"grad_norm": 4.491628170013428,
|
| 680 |
+
"learning_rate": 6.336409448665989e-07,
|
| 681 |
+
"loss": 0.4768,
|
| 682 |
+
"step": 870
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.408,
|
| 686 |
+
"grad_norm": 5.085056304931641,
|
| 687 |
+
"learning_rate": 6.031425254137223e-07,
|
| 688 |
+
"loss": 0.4539,
|
| 689 |
+
"step": 880
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.424,
|
| 693 |
+
"grad_norm": 4.508426189422607,
|
| 694 |
+
"learning_rate": 5.732105803644987e-07,
|
| 695 |
+
"loss": 0.4588,
|
| 696 |
+
"step": 890
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.44,
|
| 700 |
+
"grad_norm": 4.900942325592041,
|
| 701 |
+
"learning_rate": 5.438640153769653e-07,
|
| 702 |
+
"loss": 0.4817,
|
| 703 |
+
"step": 900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.44,
|
| 707 |
+
"eval_loss": 0.7055138945579529,
|
| 708 |
+
"eval_runtime": 9.5816,
|
| 709 |
+
"eval_samples_per_second": 208.733,
|
| 710 |
+
"eval_steps_per_second": 13.046,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.456,
|
| 715 |
+
"grad_norm": 5.169991970062256,
|
| 716 |
+
"learning_rate": 5.151213663705655e-07,
|
| 717 |
+
"loss": 0.4608,
|
| 718 |
+
"step": 910
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.472,
|
| 722 |
+
"grad_norm": 5.345458030700684,
|
| 723 |
+
"learning_rate": 4.870007878184633e-07,
|
| 724 |
+
"loss": 0.4687,
|
| 725 |
+
"step": 920
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.488,
|
| 729 |
+
"grad_norm": 5.129513263702393,
|
| 730 |
+
"learning_rate": 4.5952004128079276e-07,
|
| 731 |
+
"loss": 0.4677,
|
| 732 |
+
"step": 930
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.504,
|
| 736 |
+
"grad_norm": 4.392848491668701,
|
| 737 |
+
"learning_rate": 4.3269648418607197e-07,
|
| 738 |
+
"loss": 0.4612,
|
| 739 |
+
"step": 940
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.52,
|
| 743 |
+
"grad_norm": 4.760442733764648,
|
| 744 |
+
"learning_rate": 4.06547058867883e-07,
|
| 745 |
+
"loss": 0.4864,
|
| 746 |
+
"step": 950
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.536,
|
| 750 |
+
"grad_norm": 5.239096164703369,
|
| 751 |
+
"learning_rate": 3.8108828186372685e-07,
|
| 752 |
+
"loss": 0.4576,
|
| 753 |
+
"step": 960
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.552,
|
| 757 |
+
"grad_norm": 4.897254467010498,
|
| 758 |
+
"learning_rate": 3.56336233482828e-07,
|
| 759 |
+
"loss": 0.4518,
|
| 760 |
+
"step": 970
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.568,
|
| 764 |
+
"grad_norm": 5.0241804122924805,
|
| 765 |
+
"learning_rate": 3.32306547649465e-07,
|
| 766 |
+
"loss": 0.454,
|
| 767 |
+
"step": 980
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.584,
|
| 771 |
+
"grad_norm": 5.055388450622559,
|
| 772 |
+
"learning_rate": 3.0901440202824693e-07,
|
| 773 |
+
"loss": 0.4808,
|
| 774 |
+
"step": 990
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.6,
|
| 778 |
+
"grad_norm": 4.961503982543945,
|
| 779 |
+
"learning_rate": 2.86474508437579e-07,
|
| 780 |
+
"loss": 0.4324,
|
| 781 |
+
"step": 1000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.6,
|
| 785 |
+
"eval_loss": 0.7080456614494324,
|
| 786 |
+
"eval_runtime": 7.9696,
|
| 787 |
+
"eval_samples_per_second": 250.955,
|
| 788 |
+
"eval_steps_per_second": 15.685,
|
| 789 |
+
"step": 1000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.616,
|
| 793 |
+
"grad_norm": 4.625386714935303,
|
| 794 |
+
"learning_rate": 2.647011035573588e-07,
|
| 795 |
+
"loss": 0.4707,
|
| 796 |
+
"step": 1010
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.6320000000000001,
|
| 800 |
+
"grad_norm": 5.237055778503418,
|
| 801 |
+
"learning_rate": 2.437079399367875e-07,
|
| 802 |
+
"loss": 0.4575,
|
| 803 |
+
"step": 1020
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.6480000000000001,
|
| 807 |
+
"grad_norm": 5.412177562713623,
|
| 808 |
+
"learning_rate": 2.235082773079624e-07,
|
| 809 |
+
"loss": 0.4786,
|
| 810 |
+
"step": 1030
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 1.6640000000000001,
|
| 814 |
+
"grad_norm": 5.122972011566162,
|
| 815 |
+
"learning_rate": 2.0411487421074708e-07,
|
| 816 |
+
"loss": 0.4961,
|
| 817 |
+
"step": 1040
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.6800000000000002,
|
| 821 |
+
"grad_norm": 4.920769214630127,
|
| 822 |
+
"learning_rate": 1.8553997993420495e-07,
|
| 823 |
+
"loss": 0.4494,
|
| 824 |
+
"step": 1050
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.696,
|
| 828 |
+
"grad_norm": 4.990310192108154,
|
| 829 |
+
"learning_rate": 1.6779532677968329e-07,
|
| 830 |
+
"loss": 0.4713,
|
| 831 |
+
"step": 1060
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 1.712,
|
| 835 |
+
"grad_norm": 4.39508581161499,
|
| 836 |
+
"learning_rate": 1.508921226504434e-07,
|
| 837 |
+
"loss": 0.4633,
|
| 838 |
+
"step": 1070
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 1.728,
|
| 842 |
+
"grad_norm": 5.249621391296387,
|
| 843 |
+
"learning_rate": 1.348410439725065e-07,
|
| 844 |
+
"loss": 0.4764,
|
| 845 |
+
"step": 1080
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.744,
|
| 849 |
+
"grad_norm": 5.508373260498047,
|
| 850 |
+
"learning_rate": 1.1965222895119444e-07,
|
| 851 |
+
"loss": 0.4793,
|
| 852 |
+
"step": 1090
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.76,
|
| 856 |
+
"grad_norm": 5.088216781616211,
|
| 857 |
+
"learning_rate": 1.0533527116762298e-07,
|
| 858 |
+
"loss": 0.4693,
|
| 859 |
+
"step": 1100
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.76,
|
| 863 |
+
"eval_loss": 0.7081010937690735,
|
| 864 |
+
"eval_runtime": 7.6853,
|
| 865 |
+
"eval_samples_per_second": 260.237,
|
| 866 |
+
"eval_steps_per_second": 16.265,
|
| 867 |
+
"step": 1100
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.776,
|
| 871 |
+
"grad_norm": 5.710309028625488,
|
| 872 |
+
"learning_rate": 9.18992135191889e-08,
|
| 873 |
+
"loss": 0.4405,
|
| 874 |
+
"step": 1110
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 1.792,
|
| 878 |
+
"grad_norm": 4.6689982414245605,
|
| 879 |
+
"learning_rate": 7.935254250788366e-08,
|
| 880 |
+
"loss": 0.4484,
|
| 881 |
+
"step": 1120
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 1.808,
|
| 885 |
+
"grad_norm": 4.682315349578857,
|
| 886 |
+
"learning_rate": 6.770318288003558e-08,
|
| 887 |
+
"loss": 0.4458,
|
| 888 |
+
"step": 1130
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 1.8239999999999998,
|
| 892 |
+
"grad_norm": 5.43618106842041,
|
| 893 |
+
"learning_rate": 5.6958492620871105e-08,
|
| 894 |
+
"loss": 0.4914,
|
| 895 |
+
"step": 1140
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 1.8399999999999999,
|
| 899 |
+
"grad_norm": 4.98464822769165,
|
| 900 |
+
"learning_rate": 4.712525830705339e-08,
|
| 901 |
+
"loss": 0.4567,
|
| 902 |
+
"step": 1150
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8559999999999999,
|
| 906 |
+
"grad_norm": 4.657787799835205,
|
| 907 |
+
"learning_rate": 3.820969082013415e-08,
|
| 908 |
+
"loss": 0.452,
|
| 909 |
+
"step": 1160
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8719999999999999,
|
| 913 |
+
"grad_norm": 4.8955559730529785,
|
| 914 |
+
"learning_rate": 3.021742142362971e-08,
|
| 915 |
+
"loss": 0.463,
|
| 916 |
+
"step": 1170
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.888,
|
| 920 |
+
"grad_norm": 5.363377094268799,
|
| 921 |
+
"learning_rate": 2.3153498206192002e-08,
|
| 922 |
+
"loss": 0.4489,
|
| 923 |
+
"step": 1180
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.904,
|
| 927 |
+
"grad_norm": 4.830941677093506,
|
| 928 |
+
"learning_rate": 1.7022382893129074e-08,
|
| 929 |
+
"loss": 0.4594,
|
| 930 |
+
"step": 1190
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 1.92,
|
| 934 |
+
"grad_norm": 5.2024827003479,
|
| 935 |
+
"learning_rate": 1.1827948028283353e-08,
|
| 936 |
+
"loss": 0.4475,
|
| 937 |
+
"step": 1200
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 1.92,
|
| 941 |
+
"eval_loss": 0.707037091255188,
|
| 942 |
+
"eval_runtime": 7.654,
|
| 943 |
+
"eval_samples_per_second": 261.3,
|
| 944 |
+
"eval_steps_per_second": 16.331,
|
| 945 |
+
"step": 1200
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 1.936,
|
| 949 |
+
"grad_norm": 4.656322956085205,
|
| 950 |
+
"learning_rate": 7.57347452804974e-09,
|
| 951 |
+
"loss": 0.4583,
|
| 952 |
+
"step": 1210
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 1.952,
|
| 956 |
+
"grad_norm": 5.396835803985596,
|
| 957 |
+
"learning_rate": 4.261649609079099e-09,
|
| 958 |
+
"loss": 0.4136,
|
| 959 |
+
"step": 1220
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 1.968,
|
| 963 |
+
"grad_norm": 4.812420845031738,
|
| 964 |
+
"learning_rate": 1.8945650909737986e-09,
|
| 965 |
+
"loss": 0.4706,
|
| 966 |
+
"step": 1230
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 1.984,
|
| 970 |
+
"grad_norm": 4.871044635772705,
|
| 971 |
+
"learning_rate": 4.737160750500902e-10,
|
| 972 |
+
"loss": 0.4637,
|
| 973 |
+
"step": 1240
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.0,
|
| 977 |
+
"grad_norm": 4.772403240203857,
|
| 978 |
+
"learning_rate": 0.0,
|
| 979 |
+
"loss": 0.4707,
|
| 980 |
+
"step": 1250
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.0,
|
| 984 |
+
"step": 1250,
|
| 985 |
+
"total_flos": 4.566411071245517e+16,
|
| 986 |
+
"train_loss": 0.5696872653961181,
|
| 987 |
+
"train_runtime": 1519.873,
|
| 988 |
+
"train_samples_per_second": 13.159,
|
| 989 |
+
"train_steps_per_second": 0.822
|
| 990 |
}
|
| 991 |
],
|
| 992 |
+
"logging_steps": 10,
|
| 993 |
+
"max_steps": 1250,
|
| 994 |
"num_input_tokens_seen": 0,
|
| 995 |
+
"num_train_epochs": 2,
|
| 996 |
"save_steps": 500,
|
| 997 |
"stateful_callbacks": {
|
| 998 |
"TrainerControl": {
|
|
|
|
| 1006 |
"attributes": {}
|
| 1007 |
}
|
| 1008 |
},
|
| 1009 |
+
"total_flos": 4.566411071245517e+16,
|
| 1010 |
"train_batch_size": 2,
|
| 1011 |
"trial_name": null,
|
| 1012 |
"trial_params": null
|