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 +465 -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: oh_teknium_scaling_down_random_0.6
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_teknium_scaling_down_random_0.6
|
| 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: 0.5534
|
| 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: oh_teknium_scaling_down_random_0.6
|
|
|
|
| 16 |
|
| 17 |
# oh_teknium_scaling_down_random_0.6
|
| 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/oh_teknium_scaling_down_random_0.6 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5534
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9941520467836256,
|
| 3 |
+
"eval_loss": 0.5534155368804932,
|
| 4 |
+
"eval_runtime": 131.4214,
|
| 5 |
+
"eval_samples_per_second": 39.446,
|
| 6 |
+
"eval_steps_per_second": 0.616,
|
| 7 |
+
"total_flos": 964612073717760.0,
|
| 8 |
+
"train_loss": 0.5243237420088716,
|
| 9 |
+
"train_runtime": 19210.653,
|
| 10 |
+
"train_samples_per_second": 15.379,
|
| 11 |
+
"train_steps_per_second": 0.03
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9941520467836256,
|
| 3 |
+
"eval_loss": 0.5534155368804932,
|
| 4 |
+
"eval_runtime": 131.4214,
|
| 5 |
+
"eval_samples_per_second": 39.446,
|
| 6 |
+
"eval_steps_per_second": 0.616
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9941520467836256,
|
| 3 |
+
"total_flos": 964612073717760.0,
|
| 4 |
+
"train_loss": 0.5243237420088716,
|
| 5 |
+
"train_runtime": 19210.653,
|
| 6 |
+
"train_samples_per_second": 15.379,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,465 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9941520467836256,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 576,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05198180636777128,
|
| 13 |
+
"grad_norm": 3.0779116014443817,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7363,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.10396361273554255,
|
| 20 |
+
"grad_norm": 0.8378025538905853,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6574,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15594541910331383,
|
| 27 |
+
"grad_norm": 1.0326593461431761,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6354,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2079272254710851,
|
| 34 |
+
"grad_norm": 0.8798298291096642,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6258,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2599090318388564,
|
| 41 |
+
"grad_norm": 0.7652583221522038,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6081,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.31189083820662766,
|
| 48 |
+
"grad_norm": 0.8092758063286689,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.5957,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.36387264457439894,
|
| 55 |
+
"grad_norm": 0.6377593975168986,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.5951,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4158544509421702,
|
| 62 |
+
"grad_norm": 0.8604320761158406,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.5853,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.4678362573099415,
|
| 69 |
+
"grad_norm": 0.5659478868871771,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.5736,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5198180636777128,
|
| 76 |
+
"grad_norm": 0.6666095975840195,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.5681,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.571799870045484,
|
| 83 |
+
"grad_norm": 0.7108362162785513,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.5752,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.6237816764132553,
|
| 90 |
+
"grad_norm": 0.5055975881819947,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.5661,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6757634827810266,
|
| 97 |
+
"grad_norm": 0.5122594507537827,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.5724,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.7277452891487979,
|
| 104 |
+
"grad_norm": 0.6645034900787495,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.5642,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.7797270955165692,
|
| 111 |
+
"grad_norm": 0.4688383993640717,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.563,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.8317089018843404,
|
| 118 |
+
"grad_norm": 0.5311685960170432,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.5636,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.8836907082521117,
|
| 125 |
+
"grad_norm": 0.5275216737667017,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.5622,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.935672514619883,
|
| 132 |
+
"grad_norm": 0.49741573059254773,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.5604,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.9876543209876543,
|
| 139 |
+
"grad_norm": 0.6752670080708038,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5606,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9980506822612085,
|
| 146 |
+
"eval_loss": 0.5604754090309143,
|
| 147 |
+
"eval_runtime": 134.8008,
|
| 148 |
+
"eval_samples_per_second": 38.457,
|
| 149 |
+
"eval_steps_per_second": 0.601,
|
| 150 |
+
"step": 192
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.0396361273554255,
|
| 154 |
+
"grad_norm": 0.4942854982934459,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.5581,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.0916179337231968,
|
| 161 |
+
"grad_norm": 0.47133396673557604,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.5131,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.143599740090968,
|
| 168 |
+
"grad_norm": 0.6706210343356649,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.5108,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.1955815464587394,
|
| 175 |
+
"grad_norm": 0.4990475384686441,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.5127,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.2475633528265107,
|
| 182 |
+
"grad_norm": 0.6948432011959483,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.516,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.299545159194282,
|
| 189 |
+
"grad_norm": 0.4617478562618637,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.5124,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.3515269655620532,
|
| 196 |
+
"grad_norm": 0.40496266862562025,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.5169,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.4035087719298245,
|
| 203 |
+
"grad_norm": 0.4441934839800834,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.5153,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.4554905782975958,
|
| 210 |
+
"grad_norm": 0.5551575728453572,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.5153,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.507472384665367,
|
| 217 |
+
"grad_norm": 0.6109195782274622,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.5095,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.5594541910331383,
|
| 224 |
+
"grad_norm": 0.501987852874291,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.5134,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.6114359974009096,
|
| 231 |
+
"grad_norm": 0.42511170270357557,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5165,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.6634178037686809,
|
| 238 |
+
"grad_norm": 0.49289651016151736,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5096,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.7153996101364521,
|
| 245 |
+
"grad_norm": 0.4308251737050951,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5138,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.7673814165042234,
|
| 252 |
+
"grad_norm": 0.5157909639859894,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.5171,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.8193632228719947,
|
| 259 |
+
"grad_norm": 0.5188996520268809,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5038,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.871345029239766,
|
| 266 |
+
"grad_norm": 0.4609663557546815,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5135,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.9233268356075373,
|
| 273 |
+
"grad_norm": 0.5130233449439461,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.4969,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.9753086419753085,
|
| 280 |
+
"grad_norm": 0.47208230393580913,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5116,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.996101364522417,
|
| 287 |
+
"eval_loss": 0.5507273077964783,
|
| 288 |
+
"eval_runtime": 131.5009,
|
| 289 |
+
"eval_samples_per_second": 39.422,
|
| 290 |
+
"eval_steps_per_second": 0.616,
|
| 291 |
+
"step": 384
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.02729044834308,
|
| 295 |
+
"grad_norm": 0.6127372201164472,
|
| 296 |
+
"learning_rate": 5e-06,
|
| 297 |
+
"loss": 0.5155,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.079272254710851,
|
| 302 |
+
"grad_norm": 0.7411594983025063,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.4578,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.1312540610786224,
|
| 309 |
+
"grad_norm": 0.7428780446595316,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.4704,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.1832358674463936,
|
| 316 |
+
"grad_norm": 0.5200617783769791,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.4637,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.235217673814165,
|
| 323 |
+
"grad_norm": 0.6498653019719489,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.4622,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.287199480181936,
|
| 330 |
+
"grad_norm": 0.46276007763796656,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.4607,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.3391812865497075,
|
| 337 |
+
"grad_norm": 0.5996673377937245,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.4622,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.3911630929174787,
|
| 344 |
+
"grad_norm": 0.49488878939584074,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.4682,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.44314489928525,
|
| 351 |
+
"grad_norm": 0.4643784900227937,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.4594,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.4951267056530213,
|
| 358 |
+
"grad_norm": 0.44786233784344404,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.4675,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.5471085120207926,
|
| 365 |
+
"grad_norm": 0.6126679422066866,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.4659,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.599090318388564,
|
| 372 |
+
"grad_norm": 0.530097732347591,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.4588,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.651072124756335,
|
| 379 |
+
"grad_norm": 0.5516810761121376,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.4634,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.7030539311241064,
|
| 386 |
+
"grad_norm": 0.5437842249870648,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.4642,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.7550357374918777,
|
| 393 |
+
"grad_norm": 0.5248669258621385,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.4658,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.807017543859649,
|
| 400 |
+
"grad_norm": 0.4756943098859436,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.4712,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.8589993502274202,
|
| 407 |
+
"grad_norm": 0.5165218818933938,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.4672,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.9109811565951915,
|
| 414 |
+
"grad_norm": 0.5304020568782208,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.4682,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.962962962962963,
|
| 421 |
+
"grad_norm": 0.5954622072015786,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.4569,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.9941520467836256,
|
| 428 |
+
"eval_loss": 0.5534155368804932,
|
| 429 |
+
"eval_runtime": 134.2301,
|
| 430 |
+
"eval_samples_per_second": 38.62,
|
| 431 |
+
"eval_steps_per_second": 0.603,
|
| 432 |
+
"step": 576
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.9941520467836256,
|
| 436 |
+
"step": 576,
|
| 437 |
+
"total_flos": 964612073717760.0,
|
| 438 |
+
"train_loss": 0.5243237420088716,
|
| 439 |
+
"train_runtime": 19210.653,
|
| 440 |
+
"train_samples_per_second": 15.379,
|
| 441 |
+
"train_steps_per_second": 0.03
|
| 442 |
+
}
|
| 443 |
+
],
|
| 444 |
+
"logging_steps": 10,
|
| 445 |
+
"max_steps": 576,
|
| 446 |
+
"num_input_tokens_seen": 0,
|
| 447 |
+
"num_train_epochs": 3,
|
| 448 |
+
"save_steps": 500,
|
| 449 |
+
"stateful_callbacks": {
|
| 450 |
+
"TrainerControl": {
|
| 451 |
+
"args": {
|
| 452 |
+
"should_epoch_stop": false,
|
| 453 |
+
"should_evaluate": false,
|
| 454 |
+
"should_log": false,
|
| 455 |
+
"should_save": true,
|
| 456 |
+
"should_training_stop": true
|
| 457 |
+
},
|
| 458 |
+
"attributes": {}
|
| 459 |
+
}
|
| 460 |
+
},
|
| 461 |
+
"total_flos": 964612073717760.0,
|
| 462 |
+
"train_batch_size": 8,
|
| 463 |
+
"trial_name": null,
|
| 464 |
+
"trial_params": null
|
| 465 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|