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 +570 -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_v3.1_leave_out_non_v1.2_subsets
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_v3.1_leave_out_non_v1.2_subsets
|
| 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: 0.6441
|
| 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_v3.1_leave_out_non_v1.2_subsets
|
|
|
|
| 16 |
|
| 17 |
# oh_v3.1_leave_out_non_v1.2_subsets
|
| 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_v3.1_leave_out_non_v1.2_subsets dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6441
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6440598964691162,
|
| 4 |
+
"eval_runtime": 42.6485,
|
| 5 |
+
"eval_samples_per_second": 152.01,
|
| 6 |
+
"eval_steps_per_second": 0.61,
|
| 7 |
+
"total_flos": 1210214409830400.0,
|
| 8 |
+
"train_loss": 0.6203547849694723,
|
| 9 |
+
"train_runtime": 6426.9902,
|
| 10 |
+
"train_samples_per_second": 57.488,
|
| 11 |
+
"train_steps_per_second": 0.112
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6440598964691162,
|
| 4 |
+
"eval_runtime": 42.6485,
|
| 5 |
+
"eval_samples_per_second": 152.01,
|
| 6 |
+
"eval_steps_per_second": 0.61
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1210214409830400.0,
|
| 4 |
+
"train_loss": 0.6203547849694723,
|
| 5 |
+
"train_runtime": 6426.9902,
|
| 6 |
+
"train_samples_per_second": 57.488,
|
| 7 |
+
"train_steps_per_second": 0.112
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,570 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 723,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04149377593360996,
|
| 13 |
+
"grad_norm": 4.692910183081259,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.8926,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08298755186721991,
|
| 20 |
+
"grad_norm": 3.9587808997272367,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7811,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12448132780082988,
|
| 27 |
+
"grad_norm": 1.7341654533658646,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7551,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.16597510373443983,
|
| 34 |
+
"grad_norm": 1.2479554148072647,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.7351,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2074688796680498,
|
| 41 |
+
"grad_norm": 1.8273832972074038,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.7144,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24896265560165975,
|
| 48 |
+
"grad_norm": 0.8138076831404122,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.7064,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.29045643153526973,
|
| 55 |
+
"grad_norm": 1.1407262585354465,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.7033,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.33195020746887965,
|
| 62 |
+
"grad_norm": 0.8238677009194335,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.693,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.37344398340248963,
|
| 69 |
+
"grad_norm": 0.5627738490815636,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6908,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.4149377593360996,
|
| 76 |
+
"grad_norm": 0.49719078697733404,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6837,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.45643153526970953,
|
| 83 |
+
"grad_norm": 0.6797244336684127,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6781,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.4979253112033195,
|
| 90 |
+
"grad_norm": 0.9462681975906982,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6717,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5394190871369294,
|
| 97 |
+
"grad_norm": 0.8126221328532167,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6751,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5809128630705395,
|
| 104 |
+
"grad_norm": 0.6311672551359881,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6714,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6224066390041494,
|
| 111 |
+
"grad_norm": 0.5629380137819353,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6705,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.6639004149377593,
|
| 118 |
+
"grad_norm": 0.5156251933558172,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6616,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7053941908713693,
|
| 125 |
+
"grad_norm": 0.5430229315776672,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6732,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.7468879668049793,
|
| 132 |
+
"grad_norm": 0.562947435208453,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6618,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.7883817427385892,
|
| 139 |
+
"grad_norm": 0.8365952426281964,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6513,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.8298755186721992,
|
| 146 |
+
"grad_norm": 0.723509115051752,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6639,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.8713692946058091,
|
| 153 |
+
"grad_norm": 0.6979421362415176,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.659,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9128630705394191,
|
| 160 |
+
"grad_norm": 0.5360800645755232,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6478,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.9543568464730291,
|
| 167 |
+
"grad_norm": 0.5268876559917598,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.6627,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.995850622406639,
|
| 174 |
+
"grad_norm": 0.9017098344076718,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6609,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_loss": 0.648727297782898,
|
| 182 |
+
"eval_runtime": 43.6249,
|
| 183 |
+
"eval_samples_per_second": 148.608,
|
| 184 |
+
"eval_steps_per_second": 0.596,
|
| 185 |
+
"step": 241
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.037344398340249,
|
| 189 |
+
"grad_norm": 0.9774605752797116,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.6054,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.0788381742738589,
|
| 196 |
+
"grad_norm": 0.5579603987304247,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.6085,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.120331950207469,
|
| 203 |
+
"grad_norm": 0.5082526938460165,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.6121,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.161825726141079,
|
| 210 |
+
"grad_norm": 0.577873962008403,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.6039,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.2033195020746887,
|
| 217 |
+
"grad_norm": 0.8264610959213172,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.6084,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.2448132780082988,
|
| 224 |
+
"grad_norm": 0.7810472700290516,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.6058,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.2863070539419086,
|
| 231 |
+
"grad_norm": 1.0165843254642077,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.6028,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.3278008298755186,
|
| 238 |
+
"grad_norm": 0.5463319445923467,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.6052,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.3692946058091287,
|
| 245 |
+
"grad_norm": 0.5367970888984819,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.6091,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.4107883817427385,
|
| 252 |
+
"grad_norm": 0.5755875593246926,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6083,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.4522821576763485,
|
| 259 |
+
"grad_norm": 0.67471919059626,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.6038,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.4937759336099585,
|
| 266 |
+
"grad_norm": 0.7628032732436587,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.6044,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.5352697095435683,
|
| 273 |
+
"grad_norm": 0.8144526511601552,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.6185,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.5767634854771784,
|
| 280 |
+
"grad_norm": 0.5175614055983699,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.6053,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.6182572614107884,
|
| 287 |
+
"grad_norm": 0.5172905638008732,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.6068,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.6597510373443982,
|
| 294 |
+
"grad_norm": 0.5307106428613048,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.6085,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.7012448132780082,
|
| 301 |
+
"grad_norm": 0.6179023030500128,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.6091,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.7427385892116183,
|
| 308 |
+
"grad_norm": 0.7086042469109006,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.6179,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.784232365145228,
|
| 315 |
+
"grad_norm": 0.5176134123863614,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.605,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.8257261410788381,
|
| 322 |
+
"grad_norm": 0.5171649513530067,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.6034,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.8672199170124482,
|
| 329 |
+
"grad_norm": 0.5367432657084621,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.613,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.908713692946058,
|
| 336 |
+
"grad_norm": 0.48408277238761427,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.6057,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.950207468879668,
|
| 343 |
+
"grad_norm": 0.49669435241743065,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.6042,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.991701244813278,
|
| 350 |
+
"grad_norm": 0.48987580475367126,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.6004,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.0,
|
| 357 |
+
"eval_loss": 0.6395143866539001,
|
| 358 |
+
"eval_runtime": 43.8734,
|
| 359 |
+
"eval_samples_per_second": 147.766,
|
| 360 |
+
"eval_steps_per_second": 0.593,
|
| 361 |
+
"step": 482
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.033195020746888,
|
| 365 |
+
"grad_norm": 0.694640496714418,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.5694,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.074688796680498,
|
| 372 |
+
"grad_norm": 0.6268578643699597,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.5547,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.116182572614108,
|
| 379 |
+
"grad_norm": 0.5711614970904298,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.564,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.1576763485477177,
|
| 386 |
+
"grad_norm": 0.534249477570971,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.552,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.199170124481328,
|
| 393 |
+
"grad_norm": 0.5669497290253546,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.5608,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.240663900414938,
|
| 400 |
+
"grad_norm": 0.5226279216140413,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.5602,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.2821576763485476,
|
| 407 |
+
"grad_norm": 0.5391174076513888,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.5567,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.323651452282158,
|
| 414 |
+
"grad_norm": 0.6529505060498241,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.56,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.3651452282157677,
|
| 421 |
+
"grad_norm": 0.5210241275652869,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.5632,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.4066390041493775,
|
| 428 |
+
"grad_norm": 0.6756682017846208,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.5614,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.4481327800829877,
|
| 435 |
+
"grad_norm": 0.5404704000307257,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.5702,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.4896265560165975,
|
| 442 |
+
"grad_norm": 0.6697757206019171,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.5619,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.5311203319502074,
|
| 449 |
+
"grad_norm": 0.5048680170724003,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.5544,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.572614107883817,
|
| 456 |
+
"grad_norm": 0.5684690270985966,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.564,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.6141078838174274,
|
| 463 |
+
"grad_norm": 0.4900867200817702,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.5554,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.6556016597510372,
|
| 470 |
+
"grad_norm": 0.6334967179963366,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.5546,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.6970954356846475,
|
| 477 |
+
"grad_norm": 0.5548094054571096,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.5614,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.7385892116182573,
|
| 484 |
+
"grad_norm": 0.5034985984912429,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.5574,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.780082987551867,
|
| 491 |
+
"grad_norm": 0.5828333344441201,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.5529,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.821576763485477,
|
| 498 |
+
"grad_norm": 0.5919548268073235,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.5585,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.863070539419087,
|
| 505 |
+
"grad_norm": 0.5214263176098621,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.57,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.904564315352697,
|
| 512 |
+
"grad_norm": 0.5391043885537505,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.5619,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.9460580912863072,
|
| 519 |
+
"grad_norm": 0.5267641922982713,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.5614,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.987551867219917,
|
| 526 |
+
"grad_norm": 0.6315320754540618,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.556,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 3.0,
|
| 533 |
+
"eval_loss": 0.6440598964691162,
|
| 534 |
+
"eval_runtime": 42.4846,
|
| 535 |
+
"eval_samples_per_second": 152.597,
|
| 536 |
+
"eval_steps_per_second": 0.612,
|
| 537 |
+
"step": 723
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.0,
|
| 541 |
+
"step": 723,
|
| 542 |
+
"total_flos": 1210214409830400.0,
|
| 543 |
+
"train_loss": 0.6203547849694723,
|
| 544 |
+
"train_runtime": 6426.9902,
|
| 545 |
+
"train_samples_per_second": 57.488,
|
| 546 |
+
"train_steps_per_second": 0.112
|
| 547 |
+
}
|
| 548 |
+
],
|
| 549 |
+
"logging_steps": 10,
|
| 550 |
+
"max_steps": 723,
|
| 551 |
+
"num_input_tokens_seen": 0,
|
| 552 |
+
"num_train_epochs": 3,
|
| 553 |
+
"save_steps": 500,
|
| 554 |
+
"stateful_callbacks": {
|
| 555 |
+
"TrainerControl": {
|
| 556 |
+
"args": {
|
| 557 |
+
"should_epoch_stop": false,
|
| 558 |
+
"should_evaluate": false,
|
| 559 |
+
"should_log": false,
|
| 560 |
+
"should_save": true,
|
| 561 |
+
"should_training_stop": true
|
| 562 |
+
},
|
| 563 |
+
"attributes": {}
|
| 564 |
+
}
|
| 565 |
+
},
|
| 566 |
+
"total_flos": 1210214409830400.0,
|
| 567 |
+
"train_batch_size": 8,
|
| 568 |
+
"trial_name": null,
|
| 569 |
+
"trial_params": null
|
| 570 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|