End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +444 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: llama3-1_8b_physics_250000_samples
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_physics_250000_samples
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.7869
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: llama3-1_8b_physics_250000_samples
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_physics_250000_samples
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/physics_250000_samples dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7869
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.7868959903717041,
|
| 4 |
+
"eval_runtime": 18.2628,
|
| 5 |
+
"eval_samples_per_second": 269.565,
|
| 6 |
+
"eval_steps_per_second": 1.095,
|
| 7 |
+
"total_flos": 919595447746560.0,
|
| 8 |
+
"train_loss": 0.7765402672286459,
|
| 9 |
+
"train_runtime": 3626.8952,
|
| 10 |
+
"train_samples_per_second": 77.367,
|
| 11 |
+
"train_steps_per_second": 0.151
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.7868959903717041,
|
| 4 |
+
"eval_runtime": 18.2628,
|
| 5 |
+
"eval_samples_per_second": 269.565,
|
| 6 |
+
"eval_steps_per_second": 1.095
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 919595447746560.0,
|
| 4 |
+
"train_loss": 0.7765402672286459,
|
| 5 |
+
"train_runtime": 3626.8952,
|
| 6 |
+
"train_samples_per_second": 77.367,
|
| 7 |
+
"train_steps_per_second": 0.151
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,444 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 549,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0546448087431694,
|
| 13 |
+
"grad_norm": 9.216311123646195,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.0197,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1092896174863388,
|
| 20 |
+
"grad_norm": 2.84822970052699,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.9201,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.16393442622950818,
|
| 27 |
+
"grad_norm": 1.1976026595489564,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.8893,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2185792349726776,
|
| 34 |
+
"grad_norm": 0.7151750657967473,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.8686,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.273224043715847,
|
| 41 |
+
"grad_norm": 0.7178785706016054,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.8566,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.32786885245901637,
|
| 48 |
+
"grad_norm": 1.3738089444275707,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.8464,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3825136612021858,
|
| 55 |
+
"grad_norm": 0.7307703509770203,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.837,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4371584699453552,
|
| 62 |
+
"grad_norm": 0.9793900236646151,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.834,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.4918032786885246,
|
| 69 |
+
"grad_norm": 0.6074469784310406,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.8239,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.546448087431694,
|
| 76 |
+
"grad_norm": 0.7581415259721198,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.8203,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6010928961748634,
|
| 83 |
+
"grad_norm": 0.6207562995736402,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.8191,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.6557377049180327,
|
| 90 |
+
"grad_norm": 0.712118501956403,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.8148,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7103825136612022,
|
| 97 |
+
"grad_norm": 0.5406181591483699,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.816,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7650273224043715,
|
| 104 |
+
"grad_norm": 0.801157400991226,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.8084,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.819672131147541,
|
| 111 |
+
"grad_norm": 1.3907122538264838,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.8101,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.8743169398907104,
|
| 118 |
+
"grad_norm": 0.9374210060236701,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.8095,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9289617486338798,
|
| 125 |
+
"grad_norm": 0.7357026277429077,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.8056,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.9836065573770492,
|
| 132 |
+
"grad_norm": 0.6439903150890136,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.8067,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.0,
|
| 139 |
+
"eval_loss": 0.8027574419975281,
|
| 140 |
+
"eval_runtime": 18.4703,
|
| 141 |
+
"eval_samples_per_second": 266.536,
|
| 142 |
+
"eval_steps_per_second": 1.083,
|
| 143 |
+
"step": 183
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.0382513661202186,
|
| 147 |
+
"grad_norm": 0.9901943553002889,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.7796,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.092896174863388,
|
| 154 |
+
"grad_norm": 0.6647757287607619,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.7669,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.1475409836065573,
|
| 161 |
+
"grad_norm": 0.9580968284481025,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.7664,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.2021857923497268,
|
| 168 |
+
"grad_norm": 0.8889371423217748,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.7675,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.2568306010928962,
|
| 175 |
+
"grad_norm": 0.9247757328115808,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.7625,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.3114754098360657,
|
| 182 |
+
"grad_norm": 0.7030804123659421,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.7669,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.366120218579235,
|
| 189 |
+
"grad_norm": 0.6820492241392998,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.7679,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.4207650273224044,
|
| 196 |
+
"grad_norm": 0.5636718084741676,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.7606,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.4754098360655736,
|
| 203 |
+
"grad_norm": 0.639581027268557,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.7593,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.530054644808743,
|
| 210 |
+
"grad_norm": 0.6987258742060128,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.7612,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.5846994535519126,
|
| 217 |
+
"grad_norm": 0.6936443486720386,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.7582,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.639344262295082,
|
| 224 |
+
"grad_norm": 0.5505691855153351,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.7625,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.6939890710382515,
|
| 231 |
+
"grad_norm": 0.5640162883535096,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.7626,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.748633879781421,
|
| 238 |
+
"grad_norm": 0.6321872003984678,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.764,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.8032786885245902,
|
| 245 |
+
"grad_norm": 0.6338738498810821,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.762,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.8579234972677594,
|
| 252 |
+
"grad_norm": 0.5229579892525594,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.7672,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.9125683060109289,
|
| 259 |
+
"grad_norm": 0.7287452929104943,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.7615,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.9672131147540983,
|
| 266 |
+
"grad_norm": 0.7034416943673282,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.7596,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.0,
|
| 273 |
+
"eval_loss": 0.7888949513435364,
|
| 274 |
+
"eval_runtime": 18.3314,
|
| 275 |
+
"eval_samples_per_second": 268.556,
|
| 276 |
+
"eval_steps_per_second": 1.091,
|
| 277 |
+
"step": 366
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.021857923497268,
|
| 281 |
+
"grad_norm": 1.0144238515169914,
|
| 282 |
+
"learning_rate": 5e-06,
|
| 283 |
+
"loss": 0.7482,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.0765027322404372,
|
| 288 |
+
"grad_norm": 0.8603434912719397,
|
| 289 |
+
"learning_rate": 5e-06,
|
| 290 |
+
"loss": 0.7216,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.1311475409836067,
|
| 295 |
+
"grad_norm": 0.7563179754120521,
|
| 296 |
+
"learning_rate": 5e-06,
|
| 297 |
+
"loss": 0.7175,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.185792349726776,
|
| 302 |
+
"grad_norm": 0.9559125673842068,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.7222,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.240437158469945,
|
| 309 |
+
"grad_norm": 0.769800549346291,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.7184,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.2950819672131146,
|
| 316 |
+
"grad_norm": 0.6419920496502396,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.7222,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.349726775956284,
|
| 323 |
+
"grad_norm": 0.7532650987823766,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.7246,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.4043715846994536,
|
| 330 |
+
"grad_norm": 0.5920199594002061,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.7228,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.459016393442623,
|
| 337 |
+
"grad_norm": 0.6161056536206126,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.722,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.5136612021857925,
|
| 344 |
+
"grad_norm": 0.7438641211945972,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.7215,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.5683060109289615,
|
| 351 |
+
"grad_norm": 0.6857540384247945,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.7255,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.6229508196721314,
|
| 358 |
+
"grad_norm": 0.5697372445993161,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.7231,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.6775956284153004,
|
| 365 |
+
"grad_norm": 0.868972289179613,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.7238,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.73224043715847,
|
| 372 |
+
"grad_norm": 0.6159867401214144,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.7208,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.7868852459016393,
|
| 379 |
+
"grad_norm": 0.584105759494615,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.7263,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.841530054644809,
|
| 386 |
+
"grad_norm": 0.5906676150264403,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.7221,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.8961748633879782,
|
| 393 |
+
"grad_norm": 0.5994810510441251,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.7266,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.9508196721311473,
|
| 400 |
+
"grad_norm": 0.7788656644195687,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.7194,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 3.0,
|
| 407 |
+
"eval_loss": 0.7868959903717041,
|
| 408 |
+
"eval_runtime": 18.0047,
|
| 409 |
+
"eval_samples_per_second": 273.428,
|
| 410 |
+
"eval_steps_per_second": 1.111,
|
| 411 |
+
"step": 549
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 3.0,
|
| 415 |
+
"step": 549,
|
| 416 |
+
"total_flos": 919595447746560.0,
|
| 417 |
+
"train_loss": 0.7765402672286459,
|
| 418 |
+
"train_runtime": 3626.8952,
|
| 419 |
+
"train_samples_per_second": 77.367,
|
| 420 |
+
"train_steps_per_second": 0.151
|
| 421 |
+
}
|
| 422 |
+
],
|
| 423 |
+
"logging_steps": 10,
|
| 424 |
+
"max_steps": 549,
|
| 425 |
+
"num_input_tokens_seen": 0,
|
| 426 |
+
"num_train_epochs": 3,
|
| 427 |
+
"save_steps": 500,
|
| 428 |
+
"stateful_callbacks": {
|
| 429 |
+
"TrainerControl": {
|
| 430 |
+
"args": {
|
| 431 |
+
"should_epoch_stop": false,
|
| 432 |
+
"should_evaluate": false,
|
| 433 |
+
"should_log": false,
|
| 434 |
+
"should_save": true,
|
| 435 |
+
"should_training_stop": true
|
| 436 |
+
},
|
| 437 |
+
"attributes": {}
|
| 438 |
+
}
|
| 439 |
+
},
|
| 440 |
+
"total_flos": 919595447746560.0,
|
| 441 |
+
"train_batch_size": 16,
|
| 442 |
+
"trial_name": null,
|
| 443 |
+
"trial_params": null
|
| 444 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|