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 +703 -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-dcft-v1.2_no-curation_gpt-4o-mini_wo_evol_instruct
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh-dcft-v1.2_no-curation_gpt-4o-mini_wo_evol_instruct
|
| 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: 0.6284
|
| 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-dcft-v1.2_no-curation_gpt-4o-mini_wo_evol_instruct
|
|
|
|
| 16 |
|
| 17 |
# oh-dcft-v1.2_no-curation_gpt-4o-mini_wo_evol_instruct
|
| 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-dcft-v1.2_no-curation_gpt-4o-mini_wo_evol_instruct dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6284
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.628399670124054,
|
| 4 |
+
"eval_runtime": 29.4009,
|
| 5 |
+
"eval_samples_per_second": 280.298,
|
| 6 |
+
"eval_steps_per_second": 1.122,
|
| 7 |
+
"total_flos": 1537684191313920.0,
|
| 8 |
+
"train_loss": 0.6034856663290452,
|
| 9 |
+
"train_runtime": 5898.5619,
|
| 10 |
+
"train_samples_per_second": 79.628,
|
| 11 |
+
"train_steps_per_second": 0.156
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.628399670124054,
|
| 4 |
+
"eval_runtime": 29.4009,
|
| 5 |
+
"eval_samples_per_second": 280.298,
|
| 6 |
+
"eval_steps_per_second": 1.122
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1537684191313920.0,
|
| 4 |
+
"train_loss": 0.6034856663290452,
|
| 5 |
+
"train_runtime": 5898.5619,
|
| 6 |
+
"train_samples_per_second": 79.628,
|
| 7 |
+
"train_steps_per_second": 0.156
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,703 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 918,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.032679738562091505,
|
| 13 |
+
"grad_norm": 13.884983365234152,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.9132,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06535947712418301,
|
| 20 |
+
"grad_norm": 4.396307073532539,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.793,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09803921568627451,
|
| 27 |
+
"grad_norm": 1.5701715316993357,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7554,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13071895424836602,
|
| 34 |
+
"grad_norm": 1.0469284618172925,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.7308,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16339869281045752,
|
| 41 |
+
"grad_norm": 1.025316533729498,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.7074,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19607843137254902,
|
| 48 |
+
"grad_norm": 0.8007907881089358,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6935,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22875816993464052,
|
| 55 |
+
"grad_norm": 0.7336515686610362,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6863,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.26143790849673204,
|
| 62 |
+
"grad_norm": 0.6654762709275407,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6737,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.29411764705882354,
|
| 69 |
+
"grad_norm": 0.6752715837026988,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.678,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.32679738562091504,
|
| 76 |
+
"grad_norm": 0.6495908393250515,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6625,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.35947712418300654,
|
| 83 |
+
"grad_norm": 0.5513534888680728,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6728,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.39215686274509803,
|
| 90 |
+
"grad_norm": 0.5987482687514573,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6605,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.42483660130718953,
|
| 97 |
+
"grad_norm": 0.8982453869254353,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6576,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.45751633986928103,
|
| 104 |
+
"grad_norm": 0.5729452075955007,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6581,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.49019607843137253,
|
| 111 |
+
"grad_norm": 0.6356665135318063,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6488,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5228758169934641,
|
| 118 |
+
"grad_norm": 0.8565590645061701,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6611,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5555555555555556,
|
| 125 |
+
"grad_norm": 0.7114568602994797,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6448,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5882352941176471,
|
| 132 |
+
"grad_norm": 0.6611333193380935,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6478,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6209150326797386,
|
| 139 |
+
"grad_norm": 0.6896881110018678,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6495,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6535947712418301,
|
| 146 |
+
"grad_norm": 0.6559022818956797,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6501,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6862745098039216,
|
| 153 |
+
"grad_norm": 0.557250562288118,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.6462,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7189542483660131,
|
| 160 |
+
"grad_norm": 0.546010994925319,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6447,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7516339869281046,
|
| 167 |
+
"grad_norm": 0.6551946905014553,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.6441,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7843137254901961,
|
| 174 |
+
"grad_norm": 0.7098356396395941,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6404,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8169934640522876,
|
| 181 |
+
"grad_norm": 0.6955204100907911,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.6328,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8496732026143791,
|
| 188 |
+
"grad_norm": 0.8535008368638369,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.6376,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.8823529411764706,
|
| 195 |
+
"grad_norm": 0.9070401054614784,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.639,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9150326797385621,
|
| 202 |
+
"grad_norm": 1.565789969572099,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.6359,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9477124183006536,
|
| 209 |
+
"grad_norm": 1.2559271663411875,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.634,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.9803921568627451,
|
| 216 |
+
"grad_norm": 0.7809173302366459,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.6319,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.0,
|
| 223 |
+
"eval_loss": 0.6343755125999451,
|
| 224 |
+
"eval_runtime": 30.4377,
|
| 225 |
+
"eval_samples_per_second": 270.75,
|
| 226 |
+
"eval_steps_per_second": 1.084,
|
| 227 |
+
"step": 306
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.0130718954248366,
|
| 231 |
+
"grad_norm": 0.8552427081362785,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.613,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.0457516339869282,
|
| 238 |
+
"grad_norm": 0.7483473424289742,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5907,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.0784313725490196,
|
| 245 |
+
"grad_norm": 0.954467001493486,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5879,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.1111111111111112,
|
| 252 |
+
"grad_norm": 0.5442657939315798,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.5885,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1437908496732025,
|
| 259 |
+
"grad_norm": 0.7256618392119529,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5913,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.1764705882352942,
|
| 266 |
+
"grad_norm": 0.6093376203471469,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5941,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.2091503267973855,
|
| 273 |
+
"grad_norm": 0.563669838714874,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.5906,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.2418300653594772,
|
| 280 |
+
"grad_norm": 0.6235844786429067,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5912,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.2745098039215685,
|
| 287 |
+
"grad_norm": 0.7954936323443693,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5854,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.3071895424836601,
|
| 294 |
+
"grad_norm": 0.5900417117740443,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5902,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.3398692810457518,
|
| 301 |
+
"grad_norm": 0.5649519787360927,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5929,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.3725490196078431,
|
| 308 |
+
"grad_norm": 0.5532481138751179,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.5949,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.4052287581699345,
|
| 315 |
+
"grad_norm": 0.5796446258489036,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.591,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.4379084967320261,
|
| 322 |
+
"grad_norm": 0.5180305142972244,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5955,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.4705882352941178,
|
| 329 |
+
"grad_norm": 0.5316504409753056,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5916,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.5032679738562091,
|
| 336 |
+
"grad_norm": 0.5669201774818233,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.5884,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.5359477124183005,
|
| 343 |
+
"grad_norm": 0.5474083628106629,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.5896,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.5686274509803921,
|
| 350 |
+
"grad_norm": 0.5404778686275579,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.593,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6013071895424837,
|
| 357 |
+
"grad_norm": 0.5872446910510764,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5921,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6339869281045751,
|
| 364 |
+
"grad_norm": 0.6115573952876731,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.5916,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.6666666666666665,
|
| 371 |
+
"grad_norm": 0.6081257294896645,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.5903,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.6993464052287581,
|
| 378 |
+
"grad_norm": 0.5308744191347566,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.5859,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.7320261437908497,
|
| 385 |
+
"grad_norm": 0.5659221276589558,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5916,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.7647058823529411,
|
| 392 |
+
"grad_norm": 0.5192160551754689,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.5914,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.7973856209150327,
|
| 399 |
+
"grad_norm": 0.6988165837234015,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5866,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.8300653594771243,
|
| 406 |
+
"grad_norm": 0.5487420924541063,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5884,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.8627450980392157,
|
| 413 |
+
"grad_norm": 0.5682797543154547,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5909,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.8954248366013071,
|
| 420 |
+
"grad_norm": 0.5077833002974906,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5973,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.9281045751633987,
|
| 427 |
+
"grad_norm": 0.5129649155292699,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.5906,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.9607843137254903,
|
| 434 |
+
"grad_norm": 0.6199185848699229,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.5897,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.9934640522875817,
|
| 441 |
+
"grad_norm": 0.6367349679762161,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.5863,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.0,
|
| 448 |
+
"eval_loss": 0.6249033808708191,
|
| 449 |
+
"eval_runtime": 29.8328,
|
| 450 |
+
"eval_samples_per_second": 276.24,
|
| 451 |
+
"eval_steps_per_second": 1.106,
|
| 452 |
+
"step": 612
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.026143790849673,
|
| 456 |
+
"grad_norm": 0.7411071608832002,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.5586,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.0588235294117645,
|
| 463 |
+
"grad_norm": 0.5717129772857855,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.5407,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.0915032679738563,
|
| 470 |
+
"grad_norm": 0.6346534892777344,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.5442,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.1241830065359477,
|
| 477 |
+
"grad_norm": 0.5360417918450974,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.5411,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.156862745098039,
|
| 484 |
+
"grad_norm": 0.6070614915525901,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.5495,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.189542483660131,
|
| 491 |
+
"grad_norm": 0.5484881674300732,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.5477,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.2222222222222223,
|
| 498 |
+
"grad_norm": 0.5559727309246413,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.5453,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.2549019607843137,
|
| 505 |
+
"grad_norm": 0.6149254917908307,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.5491,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.287581699346405,
|
| 512 |
+
"grad_norm": 0.7434819330252584,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.5411,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.3202614379084965,
|
| 519 |
+
"grad_norm": 0.6277100245140601,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.539,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.3529411764705883,
|
| 526 |
+
"grad_norm": 0.6071082517695824,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.546,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.3856209150326797,
|
| 533 |
+
"grad_norm": 0.5763605350200824,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.5424,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.418300653594771,
|
| 540 |
+
"grad_norm": 0.6503780611807637,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.5467,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.450980392156863,
|
| 547 |
+
"grad_norm": 0.6953693200719112,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.551,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.4836601307189543,
|
| 554 |
+
"grad_norm": 0.582324537988359,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.5454,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.5163398692810457,
|
| 561 |
+
"grad_norm": 0.5697978800379555,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.546,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.549019607843137,
|
| 568 |
+
"grad_norm": 0.588747491644582,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.5434,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.581699346405229,
|
| 575 |
+
"grad_norm": 0.5761332349094056,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.5515,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.6143790849673203,
|
| 582 |
+
"grad_norm": 0.6003606940754802,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.5513,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.6470588235294117,
|
| 589 |
+
"grad_norm": 0.5450155999410424,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.5472,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.6797385620915035,
|
| 596 |
+
"grad_norm": 0.7041726525142917,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.5496,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.712418300653595,
|
| 603 |
+
"grad_norm": 0.5257674189970212,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.5444,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.7450980392156863,
|
| 610 |
+
"grad_norm": 0.5960484874019031,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.5501,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.7777777777777777,
|
| 617 |
+
"grad_norm": 0.5362794145960937,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.5477,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.810457516339869,
|
| 624 |
+
"grad_norm": 0.5503663269672819,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.5472,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.843137254901961,
|
| 631 |
+
"grad_norm": 0.5500519539060107,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.5512,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.8758169934640523,
|
| 638 |
+
"grad_norm": 0.6512147682006916,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.5521,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.9084967320261437,
|
| 645 |
+
"grad_norm": 0.5740057468439662,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.5449,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.9411764705882355,
|
| 652 |
+
"grad_norm": 0.5364033145980043,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.5498,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.973856209150327,
|
| 659 |
+
"grad_norm": 0.5484821254955965,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.5495,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 3.0,
|
| 666 |
+
"eval_loss": 0.628399670124054,
|
| 667 |
+
"eval_runtime": 29.5588,
|
| 668 |
+
"eval_samples_per_second": 278.8,
|
| 669 |
+
"eval_steps_per_second": 1.116,
|
| 670 |
+
"step": 918
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 3.0,
|
| 674 |
+
"step": 918,
|
| 675 |
+
"total_flos": 1537684191313920.0,
|
| 676 |
+
"train_loss": 0.6034856663290452,
|
| 677 |
+
"train_runtime": 5898.5619,
|
| 678 |
+
"train_samples_per_second": 79.628,
|
| 679 |
+
"train_steps_per_second": 0.156
|
| 680 |
+
}
|
| 681 |
+
],
|
| 682 |
+
"logging_steps": 10,
|
| 683 |
+
"max_steps": 918,
|
| 684 |
+
"num_input_tokens_seen": 0,
|
| 685 |
+
"num_train_epochs": 3,
|
| 686 |
+
"save_steps": 500,
|
| 687 |
+
"stateful_callbacks": {
|
| 688 |
+
"TrainerControl": {
|
| 689 |
+
"args": {
|
| 690 |
+
"should_epoch_stop": false,
|
| 691 |
+
"should_evaluate": false,
|
| 692 |
+
"should_log": false,
|
| 693 |
+
"should_save": true,
|
| 694 |
+
"should_training_stop": true
|
| 695 |
+
},
|
| 696 |
+
"attributes": {}
|
| 697 |
+
}
|
| 698 |
+
},
|
| 699 |
+
"total_flos": 1537684191313920.0,
|
| 700 |
+
"train_batch_size": 16,
|
| 701 |
+
"trial_name": null,
|
| 702 |
+
"trial_params": null
|
| 703 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|