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 +429 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -3,6 +3,7 @@ license: llama2
|
|
| 3 |
base_model: meta-llama/Llama-2-7b-chat-hf
|
| 4 |
tags:
|
| 5 |
- llama-factory
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: prec_240807_llama2
|
|
@@ -14,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# prec_240807_llama2
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
- Loss: 0.8686
|
| 20 |
|
|
|
|
| 3 |
base_model: meta-llama/Llama-2-7b-chat-hf
|
| 4 |
tags:
|
| 5 |
- llama-factory
|
| 6 |
+
- full
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: prec_240807_llama2
|
|
|
|
| 15 |
|
| 16 |
# prec_240807_llama2
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the perception_240723 dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.8686
|
| 21 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.8686431646347046,
|
| 4 |
+
"eval_runtime": 30.321,
|
| 5 |
+
"eval_samples_per_second": 6.596,
|
| 6 |
+
"eval_steps_per_second": 3.298,
|
| 7 |
+
"total_flos": 978757877760.0,
|
| 8 |
+
"train_loss": 0.9431339433458117,
|
| 9 |
+
"train_runtime": 1165.0317,
|
| 10 |
+
"train_samples_per_second": 1.545,
|
| 11 |
+
"train_steps_per_second": 0.386
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.8686431646347046,
|
| 4 |
+
"eval_runtime": 30.321,
|
| 5 |
+
"eval_samples_per_second": 6.596,
|
| 6 |
+
"eval_steps_per_second": 3.298
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 978757877760.0,
|
| 4 |
+
"train_loss": 0.9431339433458117,
|
| 5 |
+
"train_runtime": 1165.0317,
|
| 6 |
+
"train_samples_per_second": 1.545,
|
| 7 |
+
"train_steps_per_second": 0.386
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,429 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 450,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.022222222222222223,
|
| 13 |
+
"grad_norm": 10.848259398106622,
|
| 14 |
+
"learning_rate": 2.222222222222222e-06,
|
| 15 |
+
"loss": 1.4236,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.044444444444444446,
|
| 20 |
+
"grad_norm": 7.626496760912589,
|
| 21 |
+
"learning_rate": 4.444444444444444e-06,
|
| 22 |
+
"loss": 1.1482,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06666666666666667,
|
| 27 |
+
"grad_norm": 4.726077968840198,
|
| 28 |
+
"learning_rate": 6.666666666666667e-06,
|
| 29 |
+
"loss": 0.9179,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08888888888888889,
|
| 34 |
+
"grad_norm": 6.408303453875964,
|
| 35 |
+
"learning_rate": 8.888888888888888e-06,
|
| 36 |
+
"loss": 0.9949,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1111111111111111,
|
| 41 |
+
"grad_norm": 5.136859797071464,
|
| 42 |
+
"learning_rate": 9.996239762521152e-06,
|
| 43 |
+
"loss": 0.9956,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1111111111111111,
|
| 48 |
+
"eval_loss": 0.9926177859306335,
|
| 49 |
+
"eval_runtime": 30.2744,
|
| 50 |
+
"eval_samples_per_second": 6.606,
|
| 51 |
+
"eval_steps_per_second": 3.303,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.13333333333333333,
|
| 56 |
+
"grad_norm": 5.29564089911446,
|
| 57 |
+
"learning_rate": 9.966191788709716e-06,
|
| 58 |
+
"loss": 1.052,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.15555555555555556,
|
| 63 |
+
"grad_norm": 6.039906507025914,
|
| 64 |
+
"learning_rate": 9.906276553136924e-06,
|
| 65 |
+
"loss": 0.9701,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.17777777777777778,
|
| 70 |
+
"grad_norm": 5.578006926986183,
|
| 71 |
+
"learning_rate": 9.816854393079402e-06,
|
| 72 |
+
"loss": 1.0382,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.2,
|
| 77 |
+
"grad_norm": 5.9572609079254,
|
| 78 |
+
"learning_rate": 9.698463103929542e-06,
|
| 79 |
+
"loss": 0.9353,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2222222222222222,
|
| 84 |
+
"grad_norm": 4.116033363582691,
|
| 85 |
+
"learning_rate": 9.551814704830734e-06,
|
| 86 |
+
"loss": 0.9989,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.2222222222222222,
|
| 91 |
+
"eval_loss": 0.9735559225082397,
|
| 92 |
+
"eval_runtime": 30.2622,
|
| 93 |
+
"eval_samples_per_second": 6.609,
|
| 94 |
+
"eval_steps_per_second": 3.304,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.24444444444444444,
|
| 99 |
+
"grad_norm": 4.821110745184954,
|
| 100 |
+
"learning_rate": 9.377791156510456e-06,
|
| 101 |
+
"loss": 0.9845,
|
| 102 |
+
"step": 110
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.26666666666666666,
|
| 106 |
+
"grad_norm": 6.05004320615843,
|
| 107 |
+
"learning_rate": 9.177439057064684e-06,
|
| 108 |
+
"loss": 0.9919,
|
| 109 |
+
"step": 120
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.28888888888888886,
|
| 113 |
+
"grad_norm": 4.0399072625087085,
|
| 114 |
+
"learning_rate": 8.951963347593797e-06,
|
| 115 |
+
"loss": 0.9697,
|
| 116 |
+
"step": 130
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.3111111111111111,
|
| 120 |
+
"grad_norm": 3.82459541550962,
|
| 121 |
+
"learning_rate": 8.702720065545024e-06,
|
| 122 |
+
"loss": 0.9355,
|
| 123 |
+
"step": 140
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.3333333333333333,
|
| 127 |
+
"grad_norm": 5.29933243323369,
|
| 128 |
+
"learning_rate": 8.43120818934367e-06,
|
| 129 |
+
"loss": 0.9373,
|
| 130 |
+
"step": 150
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.3333333333333333,
|
| 134 |
+
"eval_loss": 0.9538707137107849,
|
| 135 |
+
"eval_runtime": 30.261,
|
| 136 |
+
"eval_samples_per_second": 6.609,
|
| 137 |
+
"eval_steps_per_second": 3.305,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.35555555555555557,
|
| 142 |
+
"grad_norm": 4.147741202519842,
|
| 143 |
+
"learning_rate": 8.139060623360494e-06,
|
| 144 |
+
"loss": 0.9219,
|
| 145 |
+
"step": 160
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.37777777777777777,
|
| 149 |
+
"grad_norm": 4.287684642338377,
|
| 150 |
+
"learning_rate": 7.828034377432694e-06,
|
| 151 |
+
"loss": 0.9378,
|
| 152 |
+
"step": 170
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.4,
|
| 156 |
+
"grad_norm": 4.574406510929369,
|
| 157 |
+
"learning_rate": 7.500000000000001e-06,
|
| 158 |
+
"loss": 1.0185,
|
| 159 |
+
"step": 180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.4222222222222222,
|
| 163 |
+
"grad_norm": 4.257888633665192,
|
| 164 |
+
"learning_rate": 7.156930328406268e-06,
|
| 165 |
+
"loss": 0.9562,
|
| 166 |
+
"step": 190
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.4444444444444444,
|
| 170 |
+
"grad_norm": 4.995664829906527,
|
| 171 |
+
"learning_rate": 6.800888624023552e-06,
|
| 172 |
+
"loss": 1.0155,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.4444444444444444,
|
| 177 |
+
"eval_loss": 0.9441996812820435,
|
| 178 |
+
"eval_runtime": 30.2632,
|
| 179 |
+
"eval_samples_per_second": 6.609,
|
| 180 |
+
"eval_steps_per_second": 3.304,
|
| 181 |
+
"step": 200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.4666666666666667,
|
| 185 |
+
"grad_norm": 3.9677551674894755,
|
| 186 |
+
"learning_rate": 6.434016163555452e-06,
|
| 187 |
+
"loss": 0.95,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.4888888888888889,
|
| 192 |
+
"grad_norm": 4.924489721660773,
|
| 193 |
+
"learning_rate": 6.058519361147055e-06,
|
| 194 |
+
"loss": 0.8331,
|
| 195 |
+
"step": 220
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.5111111111111111,
|
| 199 |
+
"grad_norm": 4.246970336514878,
|
| 200 |
+
"learning_rate": 5.6766564987506564e-06,
|
| 201 |
+
"loss": 0.9847,
|
| 202 |
+
"step": 230
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.5333333333333333,
|
| 206 |
+
"grad_norm": 3.657658508038241,
|
| 207 |
+
"learning_rate": 5.290724144552379e-06,
|
| 208 |
+
"loss": 0.9506,
|
| 209 |
+
"step": 240
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.5555555555555556,
|
| 213 |
+
"grad_norm": 4.911399491263737,
|
| 214 |
+
"learning_rate": 4.903043341140879e-06,
|
| 215 |
+
"loss": 0.9273,
|
| 216 |
+
"step": 250
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.5555555555555556,
|
| 220 |
+
"eval_loss": 0.9123120903968811,
|
| 221 |
+
"eval_runtime": 30.262,
|
| 222 |
+
"eval_samples_per_second": 6.609,
|
| 223 |
+
"eval_steps_per_second": 3.304,
|
| 224 |
+
"step": 250
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.5777777777777777,
|
| 228 |
+
"grad_norm": 5.034378422116753,
|
| 229 |
+
"learning_rate": 4.515945646484105e-06,
|
| 230 |
+
"loss": 0.932,
|
| 231 |
+
"step": 260
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.6,
|
| 235 |
+
"grad_norm": 4.1165067426028665,
|
| 236 |
+
"learning_rate": 4.131759111665349e-06,
|
| 237 |
+
"loss": 0.9178,
|
| 238 |
+
"step": 270
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.6222222222222222,
|
| 242 |
+
"grad_norm": 4.032023070997589,
|
| 243 |
+
"learning_rate": 3.752794279710094e-06,
|
| 244 |
+
"loss": 0.8897,
|
| 245 |
+
"step": 280
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.6444444444444445,
|
| 249 |
+
"grad_norm": 5.549145399063012,
|
| 250 |
+
"learning_rate": 3.3813302897083955e-06,
|
| 251 |
+
"loss": 0.9125,
|
| 252 |
+
"step": 290
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.6666666666666666,
|
| 256 |
+
"grad_norm": 5.343598334219672,
|
| 257 |
+
"learning_rate": 3.019601169804216e-06,
|
| 258 |
+
"loss": 0.8789,
|
| 259 |
+
"step": 300
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.6666666666666666,
|
| 263 |
+
"eval_loss": 0.896363377571106,
|
| 264 |
+
"eval_runtime": 30.257,
|
| 265 |
+
"eval_samples_per_second": 6.61,
|
| 266 |
+
"eval_steps_per_second": 3.305,
|
| 267 |
+
"step": 300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.6888888888888889,
|
| 271 |
+
"grad_norm": 4.470302563605123,
|
| 272 |
+
"learning_rate": 2.6697824014873076e-06,
|
| 273 |
+
"loss": 0.8929,
|
| 274 |
+
"step": 310
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.7111111111111111,
|
| 278 |
+
"grad_norm": 5.3357186364071065,
|
| 279 |
+
"learning_rate": 2.333977835991545e-06,
|
| 280 |
+
"loss": 0.8715,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.7333333333333333,
|
| 285 |
+
"grad_norm": 4.564334747579377,
|
| 286 |
+
"learning_rate": 2.0142070414860704e-06,
|
| 287 |
+
"loss": 0.8109,
|
| 288 |
+
"step": 330
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.7555555555555555,
|
| 292 |
+
"grad_norm": 4.12693042320115,
|
| 293 |
+
"learning_rate": 1.7123931571546826e-06,
|
| 294 |
+
"loss": 0.8895,
|
| 295 |
+
"step": 340
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.7777777777777778,
|
| 299 |
+
"grad_norm": 4.273330256106183,
|
| 300 |
+
"learning_rate": 1.4303513272105057e-06,
|
| 301 |
+
"loss": 0.871,
|
| 302 |
+
"step": 350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.7777777777777778,
|
| 306 |
+
"eval_loss": 0.8800399303436279,
|
| 307 |
+
"eval_runtime": 30.2606,
|
| 308 |
+
"eval_samples_per_second": 6.609,
|
| 309 |
+
"eval_steps_per_second": 3.305,
|
| 310 |
+
"step": 350
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8,
|
| 314 |
+
"grad_norm": 5.241490852606057,
|
| 315 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 316 |
+
"loss": 0.8507,
|
| 317 |
+
"step": 360
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.8222222222222222,
|
| 321 |
+
"grad_norm": 3.915759507331915,
|
| 322 |
+
"learning_rate": 9.322396486851626e-07,
|
| 323 |
+
"loss": 0.8813,
|
| 324 |
+
"step": 370
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8444444444444444,
|
| 328 |
+
"grad_norm": 3.5120450821860683,
|
| 329 |
+
"learning_rate": 7.191655023486682e-07,
|
| 330 |
+
"loss": 0.8444,
|
| 331 |
+
"step": 380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8666666666666667,
|
| 335 |
+
"grad_norm": 3.836463091674284,
|
| 336 |
+
"learning_rate": 5.318367983829393e-07,
|
| 337 |
+
"loss": 0.897,
|
| 338 |
+
"step": 390
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.8888888888888888,
|
| 342 |
+
"grad_norm": 4.132305775393609,
|
| 343 |
+
"learning_rate": 3.7138015365554834e-07,
|
| 344 |
+
"loss": 0.8756,
|
| 345 |
+
"step": 400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8888888888888888,
|
| 349 |
+
"eval_loss": 0.8700103759765625,
|
| 350 |
+
"eval_runtime": 30.2578,
|
| 351 |
+
"eval_samples_per_second": 6.61,
|
| 352 |
+
"eval_steps_per_second": 3.305,
|
| 353 |
+
"step": 400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9111111111111111,
|
| 357 |
+
"grad_norm": 3.412033233075354,
|
| 358 |
+
"learning_rate": 2.3876057330792344e-07,
|
| 359 |
+
"loss": 0.8607,
|
| 360 |
+
"step": 410
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9333333333333333,
|
| 364 |
+
"grad_norm": 4.485999753710129,
|
| 365 |
+
"learning_rate": 1.3477564710088097e-07,
|
| 366 |
+
"loss": 0.9037,
|
| 367 |
+
"step": 420
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9555555555555556,
|
| 371 |
+
"grad_norm": 3.873314817585464,
|
| 372 |
+
"learning_rate": 6.005075261595495e-08,
|
| 373 |
+
"loss": 0.8203,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9777777777777777,
|
| 378 |
+
"grad_norm": 4.777061740713505,
|
| 379 |
+
"learning_rate": 1.5035294161039882e-08,
|
| 380 |
+
"loss": 0.9092,
|
| 381 |
+
"step": 440
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.0,
|
| 385 |
+
"grad_norm": 4.193424030667171,
|
| 386 |
+
"learning_rate": 0.0,
|
| 387 |
+
"loss": 0.9423,
|
| 388 |
+
"step": 450
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"eval_loss": 0.8686431646347046,
|
| 393 |
+
"eval_runtime": 30.2615,
|
| 394 |
+
"eval_samples_per_second": 6.609,
|
| 395 |
+
"eval_steps_per_second": 3.305,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.0,
|
| 400 |
+
"step": 450,
|
| 401 |
+
"total_flos": 978757877760.0,
|
| 402 |
+
"train_loss": 0.9431339433458117,
|
| 403 |
+
"train_runtime": 1165.0317,
|
| 404 |
+
"train_samples_per_second": 1.545,
|
| 405 |
+
"train_steps_per_second": 0.386
|
| 406 |
+
}
|
| 407 |
+
],
|
| 408 |
+
"logging_steps": 10,
|
| 409 |
+
"max_steps": 450,
|
| 410 |
+
"num_input_tokens_seen": 0,
|
| 411 |
+
"num_train_epochs": 1,
|
| 412 |
+
"save_steps": 500,
|
| 413 |
+
"stateful_callbacks": {
|
| 414 |
+
"TrainerControl": {
|
| 415 |
+
"args": {
|
| 416 |
+
"should_epoch_stop": false,
|
| 417 |
+
"should_evaluate": false,
|
| 418 |
+
"should_log": false,
|
| 419 |
+
"should_save": false,
|
| 420 |
+
"should_training_stop": false
|
| 421 |
+
},
|
| 422 |
+
"attributes": {}
|
| 423 |
+
}
|
| 424 |
+
},
|
| 425 |
+
"total_flos": 978757877760.0,
|
| 426 |
+
"train_batch_size": 1,
|
| 427 |
+
"trial_name": null,
|
| 428 |
+
"trial_params": null
|
| 429 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|