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 +535 -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_webinstruct_original_700k
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_webinstruct_original_700k
|
| 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: 0.3991
|
| 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_webinstruct_original_700k
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_webinstruct_original_700k
|
| 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/webinstruct_original_700k dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.3991
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3991487920284271,
|
| 4 |
+
"eval_runtime": 21.3823,
|
| 5 |
+
"eval_samples_per_second": 284.254,
|
| 6 |
+
"eval_steps_per_second": 1.122,
|
| 7 |
+
"total_flos": 1135675252408320.0,
|
| 8 |
+
"train_loss": 0.38758731908151184,
|
| 9 |
+
"train_runtime": 4370.1856,
|
| 10 |
+
"train_samples_per_second": 79.274,
|
| 11 |
+
"train_steps_per_second": 0.155
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3991487920284271,
|
| 4 |
+
"eval_runtime": 21.3823,
|
| 5 |
+
"eval_samples_per_second": 284.254,
|
| 6 |
+
"eval_steps_per_second": 1.122
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1135675252408320.0,
|
| 4 |
+
"train_loss": 0.38758731908151184,
|
| 5 |
+
"train_runtime": 4370.1856,
|
| 6 |
+
"train_samples_per_second": 79.274,
|
| 7 |
+
"train_steps_per_second": 0.155
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,535 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 678,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04424778761061947,
|
| 13 |
+
"grad_norm": 6.127150945779567,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.5902,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08849557522123894,
|
| 20 |
+
"grad_norm": 5.74329347119261,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.5179,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.13274336283185842,
|
| 27 |
+
"grad_norm": 0.7209844282034923,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.4956,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.17699115044247787,
|
| 34 |
+
"grad_norm": 0.8838428515444021,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.4693,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.22123893805309736,
|
| 41 |
+
"grad_norm": 1.047241933895238,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.4542,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.26548672566371684,
|
| 48 |
+
"grad_norm": 0.6249873423196232,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.4507,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.30973451327433627,
|
| 55 |
+
"grad_norm": 0.6688340399513245,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.4422,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.35398230088495575,
|
| 62 |
+
"grad_norm": 0.5938577799260061,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.4371,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.39823008849557523,
|
| 69 |
+
"grad_norm": 0.39377946979563905,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.4288,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.4424778761061947,
|
| 76 |
+
"grad_norm": 0.3494222859720882,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.4274,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.48672566371681414,
|
| 83 |
+
"grad_norm": 0.37585161708391424,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.4204,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.5309734513274337,
|
| 90 |
+
"grad_norm": 0.41629482517793825,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.4204,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5752212389380531,
|
| 97 |
+
"grad_norm": 0.476310089674328,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.4205,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.6194690265486725,
|
| 104 |
+
"grad_norm": 0.33466258836176344,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.4189,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6637168141592921,
|
| 111 |
+
"grad_norm": 0.4017367648814896,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.4153,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7079646017699115,
|
| 118 |
+
"grad_norm": 0.4216322560941302,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.4127,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7522123893805309,
|
| 125 |
+
"grad_norm": 0.4974612613694754,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.4171,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.7964601769911505,
|
| 132 |
+
"grad_norm": 0.4002778638113851,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.4136,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.8407079646017699,
|
| 139 |
+
"grad_norm": 0.38244063424271957,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.4141,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.8849557522123894,
|
| 146 |
+
"grad_norm": 0.6422183389700243,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.4047,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9292035398230089,
|
| 153 |
+
"grad_norm": 0.47781718318215916,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.4114,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9734513274336283,
|
| 160 |
+
"grad_norm": 0.458646597184022,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.4066,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.0,
|
| 167 |
+
"eval_loss": 0.40742945671081543,
|
| 168 |
+
"eval_runtime": 21.8393,
|
| 169 |
+
"eval_samples_per_second": 278.306,
|
| 170 |
+
"eval_steps_per_second": 1.099,
|
| 171 |
+
"step": 226
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.0176991150442478,
|
| 175 |
+
"grad_norm": 0.40450956449919656,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.3954,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.0619469026548674,
|
| 182 |
+
"grad_norm": 0.33840868352830317,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.3815,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.1061946902654867,
|
| 189 |
+
"grad_norm": 0.42957928697241526,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.376,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.1504424778761062,
|
| 196 |
+
"grad_norm": 0.44727075308000447,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.3824,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.1946902654867257,
|
| 203 |
+
"grad_norm": 0.3832458382206874,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.3799,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.238938053097345,
|
| 210 |
+
"grad_norm": 0.46565842447548633,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.3763,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.2831858407079646,
|
| 217 |
+
"grad_norm": 0.3663739624439434,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.3813,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.3274336283185841,
|
| 224 |
+
"grad_norm": 0.38686016936047624,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.3825,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.3716814159292037,
|
| 231 |
+
"grad_norm": 0.4469971428096629,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.3797,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.415929203539823,
|
| 238 |
+
"grad_norm": 0.3943929709248344,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.3771,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4601769911504425,
|
| 245 |
+
"grad_norm": 0.5005987203493157,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.3797,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.504424778761062,
|
| 252 |
+
"grad_norm": 0.4398991459645874,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.3782,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.5486725663716814,
|
| 259 |
+
"grad_norm": 0.34410767390147695,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.3781,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.592920353982301,
|
| 266 |
+
"grad_norm": 0.3960889669303223,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.3749,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.6371681415929205,
|
| 273 |
+
"grad_norm": 0.4269177410578348,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.3774,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.6814159292035398,
|
| 280 |
+
"grad_norm": 0.5155261299248002,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.3804,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.7256637168141593,
|
| 287 |
+
"grad_norm": 0.43969372154192843,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.378,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.7699115044247788,
|
| 294 |
+
"grad_norm": 0.4476097491183775,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.3755,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.8141592920353982,
|
| 301 |
+
"grad_norm": 0.38686563229116155,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.3801,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.8584070796460177,
|
| 308 |
+
"grad_norm": 0.35475500773398083,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.3737,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.9026548672566372,
|
| 315 |
+
"grad_norm": 0.3300920550320411,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.377,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.9469026548672566,
|
| 322 |
+
"grad_norm": 0.37302140942747203,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.372,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.991150442477876,
|
| 329 |
+
"grad_norm": 0.3842303488060595,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.3734,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.0,
|
| 336 |
+
"eval_loss": 0.39825692772865295,
|
| 337 |
+
"eval_runtime": 21.4661,
|
| 338 |
+
"eval_samples_per_second": 283.144,
|
| 339 |
+
"eval_steps_per_second": 1.118,
|
| 340 |
+
"step": 452
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.0353982300884956,
|
| 344 |
+
"grad_norm": 0.48990695812840507,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.3513,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.079646017699115,
|
| 351 |
+
"grad_norm": 0.45716992449403504,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.3424,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.1238938053097347,
|
| 358 |
+
"grad_norm": 0.4089130956180029,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.3436,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.168141592920354,
|
| 365 |
+
"grad_norm": 0.426162580374922,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.3476,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.2123893805309733,
|
| 372 |
+
"grad_norm": 0.3887768051888474,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.3456,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.256637168141593,
|
| 379 |
+
"grad_norm": 0.49818725927434065,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.346,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.3008849557522124,
|
| 386 |
+
"grad_norm": 0.3869718840156669,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.3458,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.3451327433628317,
|
| 393 |
+
"grad_norm": 0.4420786488564716,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.3409,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.3893805309734515,
|
| 400 |
+
"grad_norm": 0.4328163120927778,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.3487,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.433628318584071,
|
| 407 |
+
"grad_norm": 0.39197907411119853,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.3451,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.47787610619469,
|
| 414 |
+
"grad_norm": 0.36416856007458076,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.3503,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.52212389380531,
|
| 421 |
+
"grad_norm": 0.3746017341827326,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.3469,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.566371681415929,
|
| 428 |
+
"grad_norm": 0.3618786916476107,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.3464,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.6106194690265485,
|
| 435 |
+
"grad_norm": 0.41598805205539596,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.3435,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.6548672566371683,
|
| 442 |
+
"grad_norm": 0.37961766765714905,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.3486,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.6991150442477876,
|
| 449 |
+
"grad_norm": 0.39738095608916224,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.3488,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.7433628318584073,
|
| 456 |
+
"grad_norm": 0.40094307810609703,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.3459,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.7876106194690267,
|
| 463 |
+
"grad_norm": 0.355834042921306,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.3415,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.831858407079646,
|
| 470 |
+
"grad_norm": 0.3745969159488807,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.35,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.8761061946902657,
|
| 477 |
+
"grad_norm": 0.3789920652707106,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.3477,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.920353982300885,
|
| 484 |
+
"grad_norm": 0.3918177591119166,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.3477,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.9646017699115044,
|
| 491 |
+
"grad_norm": 0.33578037413287726,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.3466,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.0,
|
| 498 |
+
"eval_loss": 0.3991487920284271,
|
| 499 |
+
"eval_runtime": 21.5542,
|
| 500 |
+
"eval_samples_per_second": 281.987,
|
| 501 |
+
"eval_steps_per_second": 1.113,
|
| 502 |
+
"step": 678
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 3.0,
|
| 506 |
+
"step": 678,
|
| 507 |
+
"total_flos": 1135675252408320.0,
|
| 508 |
+
"train_loss": 0.38758731908151184,
|
| 509 |
+
"train_runtime": 4370.1856,
|
| 510 |
+
"train_samples_per_second": 79.274,
|
| 511 |
+
"train_steps_per_second": 0.155
|
| 512 |
+
}
|
| 513 |
+
],
|
| 514 |
+
"logging_steps": 10,
|
| 515 |
+
"max_steps": 678,
|
| 516 |
+
"num_input_tokens_seen": 0,
|
| 517 |
+
"num_train_epochs": 3,
|
| 518 |
+
"save_steps": 500,
|
| 519 |
+
"stateful_callbacks": {
|
| 520 |
+
"TrainerControl": {
|
| 521 |
+
"args": {
|
| 522 |
+
"should_epoch_stop": false,
|
| 523 |
+
"should_evaluate": false,
|
| 524 |
+
"should_log": false,
|
| 525 |
+
"should_save": true,
|
| 526 |
+
"should_training_stop": true
|
| 527 |
+
},
|
| 528 |
+
"attributes": {}
|
| 529 |
+
}
|
| 530 |
+
},
|
| 531 |
+
"total_flos": 1135675252408320.0,
|
| 532 |
+
"train_batch_size": 16,
|
| 533 |
+
"trial_name": null,
|
| 534 |
+
"trial_params": null
|
| 535 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|