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 +633 -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_v1.3_gpt_4o_mini
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_v1.3_gpt_4o_mini
|
| 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: 0.7332
|
| 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_v1.3_gpt_4o_mini
|
|
|
|
| 16 |
|
| 17 |
# oh_v1.3_gpt_4o_mini
|
| 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_v1.3_gpt_4o_mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7332
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9954170485792853,
|
| 3 |
+
"eval_loss": 0.7332214117050171,
|
| 4 |
+
"eval_runtime": 93.8096,
|
| 5 |
+
"eval_samples_per_second": 78.35,
|
| 6 |
+
"eval_steps_per_second": 0.618,
|
| 7 |
+
"total_flos": 1366411632967680.0,
|
| 8 |
+
"train_loss": 0.7035682309491962,
|
| 9 |
+
"train_runtime": 14220.3782,
|
| 10 |
+
"train_samples_per_second": 29.46,
|
| 11 |
+
"train_steps_per_second": 0.057
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9954170485792853,
|
| 3 |
+
"eval_loss": 0.7332214117050171,
|
| 4 |
+
"eval_runtime": 93.8096,
|
| 5 |
+
"eval_samples_per_second": 78.35,
|
| 6 |
+
"eval_steps_per_second": 0.618
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9954170485792853,
|
| 3 |
+
"total_flos": 1366411632967680.0,
|
| 4 |
+
"train_loss": 0.7035682309491962,
|
| 5 |
+
"train_runtime": 14220.3782,
|
| 6 |
+
"train_samples_per_second": 29.46,
|
| 7 |
+
"train_steps_per_second": 0.057
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,633 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9954170485792853,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 816,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03666361136571952,
|
| 13 |
+
"grad_norm": 10.920966734668648,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.033,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07332722273143905,
|
| 20 |
+
"grad_norm": 2.445980198452297,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.9011,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10999083409715857,
|
| 27 |
+
"grad_norm": 1.5424232386159482,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.8764,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1466544454628781,
|
| 34 |
+
"grad_norm": 1.1797649838046136,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.8446,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.18331805682859761,
|
| 41 |
+
"grad_norm": 1.0295589020655365,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.8204,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.21998166819431714,
|
| 48 |
+
"grad_norm": 1.2160434357225554,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.8104,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2566452795600367,
|
| 55 |
+
"grad_norm": 1.2135493715768004,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.7968,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2933088909257562,
|
| 62 |
+
"grad_norm": 0.7594488712178804,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.7836,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.32997250229147573,
|
| 69 |
+
"grad_norm": 0.8913076302913621,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.7781,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.36663611365719523,
|
| 76 |
+
"grad_norm": 1.126183659145103,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.7732,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.4032997250229148,
|
| 83 |
+
"grad_norm": 0.7476760341544976,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.7711,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.4399633363886343,
|
| 90 |
+
"grad_norm": 0.828783632948725,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.7637,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4766269477543538,
|
| 97 |
+
"grad_norm": 0.7005369874659794,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.7617,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5132905591200734,
|
| 104 |
+
"grad_norm": 0.6781356553576761,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.7562,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5499541704857929,
|
| 111 |
+
"grad_norm": 0.6643060954517749,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.7601,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5866177818515124,
|
| 118 |
+
"grad_norm": 0.654862572470797,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.7561,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6232813932172319,
|
| 125 |
+
"grad_norm": 0.7126834121476828,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.7549,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6599450045829515,
|
| 132 |
+
"grad_norm": 0.5845932549413623,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.7525,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.696608615948671,
|
| 139 |
+
"grad_norm": 0.583642927450063,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.7507,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.7332722273143905,
|
| 146 |
+
"grad_norm": 0.5759630428428489,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.7492,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.76993583868011,
|
| 153 |
+
"grad_norm": 0.597809207757354,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.7446,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.8065994500458296,
|
| 160 |
+
"grad_norm": 0.6520665055230834,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.7512,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.843263061411549,
|
| 167 |
+
"grad_norm": 0.6521761800994458,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.744,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8799266727772685,
|
| 174 |
+
"grad_norm": 0.6083361886529014,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.7431,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.916590284142988,
|
| 181 |
+
"grad_norm": 0.8966782629847545,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.7399,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.9532538955087076,
|
| 188 |
+
"grad_norm": 0.6584181334872885,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.7457,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9899175068744271,
|
| 195 |
+
"grad_norm": 0.5614900416740534,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.7434,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.997250229147571,
|
| 202 |
+
"eval_loss": 0.743977963924408,
|
| 203 |
+
"eval_runtime": 96.6447,
|
| 204 |
+
"eval_samples_per_second": 76.052,
|
| 205 |
+
"eval_steps_per_second": 0.6,
|
| 206 |
+
"step": 272
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.0284142988084326,
|
| 210 |
+
"grad_norm": 0.6377526876986616,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.7593,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.065077910174152,
|
| 217 |
+
"grad_norm": 0.8312923337011684,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.6885,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.1017415215398716,
|
| 224 |
+
"grad_norm": 0.6499984381614756,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.6893,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.138405132905591,
|
| 231 |
+
"grad_norm": 0.658519279927457,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.6868,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.1750687442713108,
|
| 238 |
+
"grad_norm": 0.6307182099292118,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.6885,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.2117323556370303,
|
| 245 |
+
"grad_norm": 0.6191143311988347,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.6871,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.2483959670027498,
|
| 252 |
+
"grad_norm": 0.6735946598593434,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6935,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.2850595783684693,
|
| 259 |
+
"grad_norm": 0.7213451984916242,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.6943,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.3217231897341888,
|
| 266 |
+
"grad_norm": 0.5841901070016948,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.6938,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.3583868010999083,
|
| 273 |
+
"grad_norm": 0.6609752377099979,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.6856,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.3950504124656278,
|
| 280 |
+
"grad_norm": 0.6004672142282963,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.69,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.4317140238313475,
|
| 287 |
+
"grad_norm": 0.7494020947088555,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.682,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.468377635197067,
|
| 294 |
+
"grad_norm": 0.6711006066177567,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.6917,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.5050412465627865,
|
| 301 |
+
"grad_norm": 0.6517430215570676,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.6871,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.541704857928506,
|
| 308 |
+
"grad_norm": 0.6180564693914907,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.6829,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.5783684692942255,
|
| 315 |
+
"grad_norm": 0.5764324092377354,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.6824,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.615032080659945,
|
| 322 |
+
"grad_norm": 0.7134204082562298,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.6822,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.6516956920256645,
|
| 329 |
+
"grad_norm": 0.7630512170385407,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.6879,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.6883593033913842,
|
| 336 |
+
"grad_norm": 0.6285437172539765,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.6804,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.7250229147571035,
|
| 343 |
+
"grad_norm": 0.5968789313484854,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.686,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.7616865261228232,
|
| 350 |
+
"grad_norm": 0.6425175740435289,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.6856,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.7983501374885427,
|
| 357 |
+
"grad_norm": 0.7614365266625939,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.6814,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.8350137488542622,
|
| 364 |
+
"grad_norm": 0.5496379357416068,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.6855,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.8716773602199817,
|
| 371 |
+
"grad_norm": 0.8494093367270151,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.6875,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.9083409715857012,
|
| 378 |
+
"grad_norm": 0.6756166103000668,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.6856,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.9450045829514209,
|
| 385 |
+
"grad_norm": 0.7228484772895967,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.6841,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.9816681943171401,
|
| 392 |
+
"grad_norm": 0.7786774729146112,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.6845,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.996333638863428,
|
| 399 |
+
"eval_loss": 0.7305116057395935,
|
| 400 |
+
"eval_runtime": 96.1532,
|
| 401 |
+
"eval_samples_per_second": 76.441,
|
| 402 |
+
"eval_steps_per_second": 0.603,
|
| 403 |
+
"step": 544
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.020164986251146,
|
| 407 |
+
"grad_norm": 1.0425759476709966,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.707,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.056828597616865,
|
| 414 |
+
"grad_norm": 0.8473344095764829,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.6313,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.093492208982585,
|
| 421 |
+
"grad_norm": 0.7205628261028438,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.6281,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.130155820348304,
|
| 428 |
+
"grad_norm": 0.6604987014823058,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.631,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.166819431714024,
|
| 435 |
+
"grad_norm": 0.6774961015973217,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.6351,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.203483043079743,
|
| 442 |
+
"grad_norm": 0.8519809292040578,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.634,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.240146654445463,
|
| 449 |
+
"grad_norm": 0.693823740633704,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.6327,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.276810265811182,
|
| 456 |
+
"grad_norm": 0.6448705487045298,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.6339,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.313473877176902,
|
| 463 |
+
"grad_norm": 0.5865817788059118,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.636,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.3501374885426216,
|
| 470 |
+
"grad_norm": 0.8116556137845999,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.6342,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.386801099908341,
|
| 477 |
+
"grad_norm": 0.6231657257473445,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.637,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.4234647112740606,
|
| 484 |
+
"grad_norm": 0.6250913266909794,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.63,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.46012832263978,
|
| 491 |
+
"grad_norm": 0.582068921531117,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.6288,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.4967919340054996,
|
| 498 |
+
"grad_norm": 0.6912367969819871,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.6381,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.5334555453712193,
|
| 505 |
+
"grad_norm": 0.7147652107920064,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.6332,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.5701191567369386,
|
| 512 |
+
"grad_norm": 0.5792260811836798,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.6351,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.606782768102658,
|
| 519 |
+
"grad_norm": 0.7963438662743851,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.6363,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.6434463794683776,
|
| 526 |
+
"grad_norm": 0.9276380358330181,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.6355,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.6801099908340973,
|
| 533 |
+
"grad_norm": 0.9313823270809661,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.6351,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.7167736021998166,
|
| 540 |
+
"grad_norm": 0.7304200587600748,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.638,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.7534372135655363,
|
| 547 |
+
"grad_norm": 0.6212966397528322,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.6388,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.7901008249312556,
|
| 554 |
+
"grad_norm": 0.6720686482466423,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.6364,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.8267644362969753,
|
| 561 |
+
"grad_norm": 0.6438467896193539,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.6421,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.863428047662695,
|
| 568 |
+
"grad_norm": 0.6043416931907646,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.6379,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.9000916590284143,
|
| 575 |
+
"grad_norm": 0.6496494693588303,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.6414,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.936755270394134,
|
| 582 |
+
"grad_norm": 0.8144443719589332,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.6361,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.9734188817598532,
|
| 589 |
+
"grad_norm": 0.7037764123768507,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.6373,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.9954170485792853,
|
| 596 |
+
"eval_loss": 0.7332214117050171,
|
| 597 |
+
"eval_runtime": 94.4197,
|
| 598 |
+
"eval_samples_per_second": 77.844,
|
| 599 |
+
"eval_steps_per_second": 0.614,
|
| 600 |
+
"step": 816
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.9954170485792853,
|
| 604 |
+
"step": 816,
|
| 605 |
+
"total_flos": 1366411632967680.0,
|
| 606 |
+
"train_loss": 0.7035682309491962,
|
| 607 |
+
"train_runtime": 14220.3782,
|
| 608 |
+
"train_samples_per_second": 29.46,
|
| 609 |
+
"train_steps_per_second": 0.057
|
| 610 |
+
}
|
| 611 |
+
],
|
| 612 |
+
"logging_steps": 10,
|
| 613 |
+
"max_steps": 816,
|
| 614 |
+
"num_input_tokens_seen": 0,
|
| 615 |
+
"num_train_epochs": 3,
|
| 616 |
+
"save_steps": 500,
|
| 617 |
+
"stateful_callbacks": {
|
| 618 |
+
"TrainerControl": {
|
| 619 |
+
"args": {
|
| 620 |
+
"should_epoch_stop": false,
|
| 621 |
+
"should_evaluate": false,
|
| 622 |
+
"should_log": false,
|
| 623 |
+
"should_save": true,
|
| 624 |
+
"should_training_stop": true
|
| 625 |
+
},
|
| 626 |
+
"attributes": {}
|
| 627 |
+
}
|
| 628 |
+
},
|
| 629 |
+
"total_flos": 1366411632967680.0,
|
| 630 |
+
"train_batch_size": 8,
|
| 631 |
+
"trial_name": null,
|
| 632 |
+
"trial_params": null
|
| 633 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|