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/Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: oh_v1_w_v3_metamath
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_v1_w_v3_metamath
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5682
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: oh_v1_w_v3_metamath
|
|
|
|
| 16 |
|
| 17 |
# oh_v1_w_v3_metamath
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/oh_v1_w_v3_metamath dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5682
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.992616899097621,
|
| 3 |
+
"eval_loss": 0.5682421326637268,
|
| 4 |
+
"eval_runtime": 163.9676,
|
| 5 |
+
"eval_samples_per_second": 50.059,
|
| 6 |
+
"eval_steps_per_second": 0.396,
|
| 7 |
+
"total_flos": 1527215208529920.0,
|
| 8 |
+
"train_loss": 0.5539279884021533,
|
| 9 |
+
"train_runtime": 27411.4767,
|
| 10 |
+
"train_samples_per_second": 17.066,
|
| 11 |
+
"train_steps_per_second": 0.033
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.992616899097621,
|
| 3 |
+
"eval_loss": 0.5682421326637268,
|
| 4 |
+
"eval_runtime": 163.9676,
|
| 5 |
+
"eval_samples_per_second": 50.059,
|
| 6 |
+
"eval_steps_per_second": 0.396
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.992616899097621,
|
| 3 |
+
"total_flos": 1527215208529920.0,
|
| 4 |
+
"train_loss": 0.5539279884021533,
|
| 5 |
+
"train_runtime": 27411.4767,
|
| 6 |
+
"train_samples_per_second": 17.066,
|
| 7 |
+
"train_steps_per_second": 0.033
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,703 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.992616899097621,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 912,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03281378178835111,
|
| 13 |
+
"grad_norm": 7.841750232611161,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.8915,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06562756357670221,
|
| 20 |
+
"grad_norm": 2.8254265829702723,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7744,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09844134536505332,
|
| 27 |
+
"grad_norm": 1.2137270648230942,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7284,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13125512715340443,
|
| 34 |
+
"grad_norm": 1.0119344856669166,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.7015,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16406890894175555,
|
| 41 |
+
"grad_norm": 1.1590120587560695,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6751,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19688269073010664,
|
| 48 |
+
"grad_norm": 1.0688354918846648,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6539,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22969647251845776,
|
| 55 |
+
"grad_norm": 0.7962599254830892,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6416,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.26251025430680885,
|
| 62 |
+
"grad_norm": 1.3016899088803453,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6401,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.29532403609515995,
|
| 69 |
+
"grad_norm": 0.8330196899066211,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.618,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3281378178835111,
|
| 76 |
+
"grad_norm": 0.667189033403046,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6196,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3609515996718622,
|
| 83 |
+
"grad_norm": 0.7289539260596677,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6297,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3937653814602133,
|
| 90 |
+
"grad_norm": 0.6409803463860579,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6216,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4265791632485644,
|
| 97 |
+
"grad_norm": 0.70735959118925,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6092,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4593929450369155,
|
| 104 |
+
"grad_norm": 0.8142813291097145,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6024,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4922067268252666,
|
| 111 |
+
"grad_norm": 0.6426092966393037,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6028,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5250205086136177,
|
| 118 |
+
"grad_norm": 0.9814251564680376,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6042,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5578342904019689,
|
| 125 |
+
"grad_norm": 0.601307551623306,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6047,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5906480721903199,
|
| 132 |
+
"grad_norm": 0.5895692038996282,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.5936,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.623461853978671,
|
| 139 |
+
"grad_norm": 0.5348923163508846,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5906,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6562756357670222,
|
| 146 |
+
"grad_norm": 0.6710759767714962,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.5863,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6890894175553732,
|
| 153 |
+
"grad_norm": 0.5530954428148116,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.5972,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7219031993437244,
|
| 160 |
+
"grad_norm": 0.6414930112608357,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.5888,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7547169811320755,
|
| 167 |
+
"grad_norm": 0.730574957468606,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.5913,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7875307629204266,
|
| 174 |
+
"grad_norm": 0.8034131557047212,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.5849,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8203445447087777,
|
| 181 |
+
"grad_norm": 0.6772004609931918,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.5904,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8531583264971287,
|
| 188 |
+
"grad_norm": 0.5963848396035898,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.5814,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.8859721082854799,
|
| 195 |
+
"grad_norm": 0.8533728487617155,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.5926,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.918785890073831,
|
| 202 |
+
"grad_norm": 0.6929371502886701,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.584,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9515996718621821,
|
| 209 |
+
"grad_norm": 1.0646930858335608,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.583,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.9844134536505332,
|
| 216 |
+
"grad_norm": 0.657886516351812,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.5822,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9975389663658737,
|
| 223 |
+
"eval_loss": 0.5811063051223755,
|
| 224 |
+
"eval_runtime": 164.1949,
|
| 225 |
+
"eval_samples_per_second": 49.989,
|
| 226 |
+
"eval_steps_per_second": 0.396,
|
| 227 |
+
"step": 304
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.0172272354388843,
|
| 231 |
+
"grad_norm": 0.9145633972949372,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5615,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.0500410172272354,
|
| 238 |
+
"grad_norm": 0.8330439338358776,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5501,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.0828547990155866,
|
| 245 |
+
"grad_norm": 0.9017808771885161,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5397,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.1156685808039377,
|
| 252 |
+
"grad_norm": 0.7658877957248926,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.543,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1484823625922886,
|
| 259 |
+
"grad_norm": 0.559631564116058,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5402,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.1812961443806398,
|
| 266 |
+
"grad_norm": 0.7555936047357931,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5424,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.214109926168991,
|
| 273 |
+
"grad_norm": 0.7751086823302814,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.5338,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.246923707957342,
|
| 280 |
+
"grad_norm": 1.1576515759835613,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5447,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.2797374897456932,
|
| 287 |
+
"grad_norm": 0.637673440804281,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5436,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.3125512715340442,
|
| 294 |
+
"grad_norm": 0.5608427930549431,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.538,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.3453650533223955,
|
| 301 |
+
"grad_norm": 0.6236715534816973,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5401,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.3781788351107465,
|
| 308 |
+
"grad_norm": 0.6834154815439046,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.5371,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.4109926168990976,
|
| 315 |
+
"grad_norm": 0.6230411253152371,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5396,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.4438063986874488,
|
| 322 |
+
"grad_norm": 0.5735284718821723,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.546,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.4766201804758,
|
| 329 |
+
"grad_norm": 0.6777341683980052,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5312,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.509433962264151,
|
| 336 |
+
"grad_norm": 0.6145373481418245,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.5363,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.542247744052502,
|
| 343 |
+
"grad_norm": 0.6691485262612896,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.5405,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.5750615258408531,
|
| 350 |
+
"grad_norm": 0.6278473705145111,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.5336,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6078753076292043,
|
| 357 |
+
"grad_norm": 0.7195303747825454,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5456,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6406890894175554,
|
| 364 |
+
"grad_norm": 0.6993265245998453,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.526,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.6735028712059066,
|
| 371 |
+
"grad_norm": 0.5569198466336033,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.5437,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.7063166529942575,
|
| 378 |
+
"grad_norm": 0.5962729727283363,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.54,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.7391304347826086,
|
| 385 |
+
"grad_norm": 0.5528562495704502,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5348,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.7719442165709598,
|
| 392 |
+
"grad_norm": 0.5117801135768556,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.5354,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.804757998359311,
|
| 399 |
+
"grad_norm": 0.6270359146501516,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5405,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.837571780147662,
|
| 406 |
+
"grad_norm": 0.49991108374112964,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5311,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.870385561936013,
|
| 413 |
+
"grad_norm": 0.8786052238787824,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5276,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.9031993437243644,
|
| 420 |
+
"grad_norm": 0.5526276905535635,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5303,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.9360131255127153,
|
| 427 |
+
"grad_norm": 0.6424736342148375,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.5378,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.9688269073010665,
|
| 434 |
+
"grad_norm": 0.4985928762767277,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.5331,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.9983593109105824,
|
| 441 |
+
"eval_loss": 0.5679268836975098,
|
| 442 |
+
"eval_runtime": 165.1141,
|
| 443 |
+
"eval_samples_per_second": 49.711,
|
| 444 |
+
"eval_steps_per_second": 0.394,
|
| 445 |
+
"step": 609
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.0016406890894176,
|
| 449 |
+
"grad_norm": 0.8535377767409578,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.5258,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.0344544708777685,
|
| 456 |
+
"grad_norm": 0.8392464205497018,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.4982,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.06726825266612,
|
| 463 |
+
"grad_norm": 0.6991396805774981,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.4856,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.100082034454471,
|
| 470 |
+
"grad_norm": 0.6363700778523905,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.49,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.132895816242822,
|
| 477 |
+
"grad_norm": 0.6689991208530507,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.49,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.165709598031173,
|
| 484 |
+
"grad_norm": 0.5725274111655418,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.4926,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.198523379819524,
|
| 491 |
+
"grad_norm": 0.8141370837443801,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.4876,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.2313371616078754,
|
| 498 |
+
"grad_norm": 0.6282724145860844,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.4878,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.2641509433962264,
|
| 505 |
+
"grad_norm": 0.6530073663418564,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.4944,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.2969647251845773,
|
| 512 |
+
"grad_norm": 0.5699233164866262,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.4966,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.3297785069729287,
|
| 519 |
+
"grad_norm": 0.610417793848035,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.49,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.3625922887612796,
|
| 526 |
+
"grad_norm": 0.6289268935164573,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.5036,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.395406070549631,
|
| 533 |
+
"grad_norm": 0.6537483086050372,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.4927,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.428219852337982,
|
| 540 |
+
"grad_norm": 0.6648368643526823,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.4983,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.4610336341263332,
|
| 547 |
+
"grad_norm": 0.6722645125843071,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.4949,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.493847415914684,
|
| 554 |
+
"grad_norm": 0.5717762614134606,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.5014,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.526661197703035,
|
| 561 |
+
"grad_norm": 0.6732475102422392,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.4978,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.5594749794913865,
|
| 568 |
+
"grad_norm": 0.6176484271038984,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.4969,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.5922887612797374,
|
| 575 |
+
"grad_norm": 0.5794899161431503,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.497,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.6251025430680883,
|
| 582 |
+
"grad_norm": 0.6254525725000455,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.4881,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.6579163248564397,
|
| 589 |
+
"grad_norm": 0.4997430504095228,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.4986,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.690730106644791,
|
| 596 |
+
"grad_norm": 0.6452212092676007,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.4937,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.723543888433142,
|
| 603 |
+
"grad_norm": 0.6248001192436421,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.5017,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.756357670221493,
|
| 610 |
+
"grad_norm": 0.5729926597313496,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.4997,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.7891714520098443,
|
| 617 |
+
"grad_norm": 0.6322042992676079,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.4975,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.821985233798195,
|
| 624 |
+
"grad_norm": 0.6001289917677007,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.4932,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.854799015586546,
|
| 631 |
+
"grad_norm": 0.6372644243152534,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.5005,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.8876127973748975,
|
| 638 |
+
"grad_norm": 0.6486605042735685,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.4996,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.9204265791632484,
|
| 645 |
+
"grad_norm": 0.5285905512046695,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.5018,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.9532403609516,
|
| 652 |
+
"grad_norm": 0.6212025899780993,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.4993,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.9860541427399507,
|
| 659 |
+
"grad_norm": 0.6141734836175889,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.4973,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.992616899097621,
|
| 666 |
+
"eval_loss": 0.5682421326637268,
|
| 667 |
+
"eval_runtime": 165.8526,
|
| 668 |
+
"eval_samples_per_second": 49.49,
|
| 669 |
+
"eval_steps_per_second": 0.392,
|
| 670 |
+
"step": 912
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.992616899097621,
|
| 674 |
+
"step": 912,
|
| 675 |
+
"total_flos": 1527215208529920.0,
|
| 676 |
+
"train_loss": 0.5539279884021533,
|
| 677 |
+
"train_runtime": 27411.4767,
|
| 678 |
+
"train_samples_per_second": 17.066,
|
| 679 |
+
"train_steps_per_second": 0.033
|
| 680 |
+
}
|
| 681 |
+
],
|
| 682 |
+
"logging_steps": 10,
|
| 683 |
+
"max_steps": 912,
|
| 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": 1527215208529920.0,
|
| 700 |
+
"train_batch_size": 8,
|
| 701 |
+
"trial_name": null,
|
| 702 |
+
"trial_params": null
|
| 703 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|