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 +647 -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: llama3-1_8b_physics_375000_samples
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_physics_375000_samples
|
| 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: 0.7752
|
| 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: llama3-1_8b_physics_375000_samples
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_physics_375000_samples
|
| 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/physics_375000_samples dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7752
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.7752296328544617,
|
| 4 |
+
"eval_runtime": 27.3716,
|
| 5 |
+
"eval_samples_per_second": 273.423,
|
| 6 |
+
"eval_steps_per_second": 1.096,
|
| 7 |
+
"total_flos": 1396981062696960.0,
|
| 8 |
+
"train_loss": 0.7675551453368555,
|
| 9 |
+
"train_runtime": 5571.5313,
|
| 10 |
+
"train_samples_per_second": 76.563,
|
| 11 |
+
"train_steps_per_second": 0.15
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.7752296328544617,
|
| 4 |
+
"eval_runtime": 27.3716,
|
| 5 |
+
"eval_samples_per_second": 273.423,
|
| 6 |
+
"eval_steps_per_second": 1.096
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1396981062696960.0,
|
| 4 |
+
"train_loss": 0.7675551453368555,
|
| 5 |
+
"train_runtime": 5571.5313,
|
| 6 |
+
"train_samples_per_second": 76.563,
|
| 7 |
+
"train_steps_per_second": 0.15
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,647 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 834,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03597122302158273,
|
| 13 |
+
"grad_norm": 26.93948859971876,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.0175,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07194244604316546,
|
| 20 |
+
"grad_norm": 2.6743423167480063,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.9337,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1079136690647482,
|
| 27 |
+
"grad_norm": 1.100805597904231,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.8923,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14388489208633093,
|
| 34 |
+
"grad_norm": 0.8355539701078896,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.8673,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.17985611510791366,
|
| 41 |
+
"grad_norm": 0.7157506047100403,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.8553,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2158273381294964,
|
| 48 |
+
"grad_norm": 0.9806631521043339,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.8492,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2517985611510791,
|
| 55 |
+
"grad_norm": 0.8360835611944488,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.8382,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.28776978417266186,
|
| 62 |
+
"grad_norm": 0.7078472519601653,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.8318,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3237410071942446,
|
| 69 |
+
"grad_norm": 0.6255785562847258,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.825,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3597122302158273,
|
| 76 |
+
"grad_norm": 0.6950072028339258,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.8225,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.39568345323741005,
|
| 83 |
+
"grad_norm": 0.622757689781733,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.8165,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.4316546762589928,
|
| 90 |
+
"grad_norm": 0.6855173384055511,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.8162,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4676258992805755,
|
| 97 |
+
"grad_norm": 0.555459004966806,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.8141,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5035971223021583,
|
| 104 |
+
"grad_norm": 0.7189252900166325,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.8113,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.539568345323741,
|
| 111 |
+
"grad_norm": 0.8411135438726722,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.8069,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5755395683453237,
|
| 118 |
+
"grad_norm": 0.9141854769887011,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.8087,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6115107913669064,
|
| 125 |
+
"grad_norm": 0.6527584548807389,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.8048,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6474820143884892,
|
| 132 |
+
"grad_norm": 0.6986581112545092,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.8051,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6834532374100719,
|
| 139 |
+
"grad_norm": 0.6094857952430536,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.8044,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.7194244604316546,
|
| 146 |
+
"grad_norm": 0.74096920276776,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.7989,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.7553956834532374,
|
| 153 |
+
"grad_norm": 0.6584952886572538,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.8025,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7913669064748201,
|
| 160 |
+
"grad_norm": 0.5838446606699556,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.7988,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.8273381294964028,
|
| 167 |
+
"grad_norm": 0.5916175411049406,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.7985,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8633093525179856,
|
| 174 |
+
"grad_norm": 0.626471567693148,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.7973,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8992805755395683,
|
| 181 |
+
"grad_norm": 0.6338741269795162,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.7933,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.935251798561151,
|
| 188 |
+
"grad_norm": 0.8343555675066444,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.7969,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9712230215827338,
|
| 195 |
+
"grad_norm": 0.6221641429373133,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.7933,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.0,
|
| 202 |
+
"eval_loss": 0.7923575043678284,
|
| 203 |
+
"eval_runtime": 27.9533,
|
| 204 |
+
"eval_samples_per_second": 267.732,
|
| 205 |
+
"eval_steps_per_second": 1.073,
|
| 206 |
+
"step": 278
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.0071942446043165,
|
| 210 |
+
"grad_norm": 0.8944971285319924,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.7823,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.0431654676258992,
|
| 217 |
+
"grad_norm": 0.7668083853056575,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.7574,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.079136690647482,
|
| 224 |
+
"grad_norm": 0.6176816592509634,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.7529,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.1151079136690647,
|
| 231 |
+
"grad_norm": 0.6475301176330789,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.7558,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.1510791366906474,
|
| 238 |
+
"grad_norm": 0.5811910989874788,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.7623,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.1870503597122302,
|
| 245 |
+
"grad_norm": 0.6269454462814978,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.7601,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.223021582733813,
|
| 252 |
+
"grad_norm": 0.5423886247053047,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.7535,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.2589928057553956,
|
| 259 |
+
"grad_norm": 0.6670401432003603,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.757,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.2949640287769784,
|
| 266 |
+
"grad_norm": 0.7095322132659916,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.759,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.330935251798561,
|
| 273 |
+
"grad_norm": 0.6870367808903867,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.7567,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.3669064748201438,
|
| 280 |
+
"grad_norm": 0.6640094117573664,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.7592,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.4028776978417266,
|
| 287 |
+
"grad_norm": 0.5994950619117767,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.7529,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.4388489208633093,
|
| 294 |
+
"grad_norm": 0.7392872817621052,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.7554,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.474820143884892,
|
| 301 |
+
"grad_norm": 0.5656749568866071,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.7547,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.5107913669064748,
|
| 308 |
+
"grad_norm": 0.921484641426356,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.7532,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.5467625899280577,
|
| 315 |
+
"grad_norm": 0.540059029380678,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.7585,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.5827338129496402,
|
| 322 |
+
"grad_norm": 0.6558652758296812,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.7515,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.6187050359712232,
|
| 329 |
+
"grad_norm": 0.57268163367781,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.7562,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.6546762589928057,
|
| 336 |
+
"grad_norm": 0.5407189047091853,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.7559,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.6906474820143886,
|
| 343 |
+
"grad_norm": 0.6077940984618293,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.757,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.7266187050359711,
|
| 350 |
+
"grad_norm": 1.001124812241379,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.7552,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.762589928057554,
|
| 357 |
+
"grad_norm": 0.6254013722291123,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.753,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.7985611510791366,
|
| 364 |
+
"grad_norm": 0.5767617312575639,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.7594,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.8345323741007196,
|
| 371 |
+
"grad_norm": 0.665915353902276,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.7554,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.870503597122302,
|
| 378 |
+
"grad_norm": 0.5596777388150926,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.7537,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.906474820143885,
|
| 385 |
+
"grad_norm": 0.5547398560915929,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.7555,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.9424460431654675,
|
| 392 |
+
"grad_norm": 0.5874602156110944,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.7509,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.9784172661870505,
|
| 399 |
+
"grad_norm": 0.6369533697170318,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.7503,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 2.0,
|
| 406 |
+
"eval_loss": 0.7788412570953369,
|
| 407 |
+
"eval_runtime": 27.8988,
|
| 408 |
+
"eval_samples_per_second": 268.255,
|
| 409 |
+
"eval_steps_per_second": 1.075,
|
| 410 |
+
"step": 556
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.014388489208633,
|
| 414 |
+
"grad_norm": 1.0929207520027995,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.735,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.050359712230216,
|
| 421 |
+
"grad_norm": 0.687310495052166,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.7131,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.0863309352517985,
|
| 428 |
+
"grad_norm": 0.6848749958758751,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.7129,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.1223021582733814,
|
| 435 |
+
"grad_norm": 0.9700661070159223,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.7154,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.158273381294964,
|
| 442 |
+
"grad_norm": 0.7429316335562708,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.7163,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.194244604316547,
|
| 449 |
+
"grad_norm": 0.5731198010767242,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.7197,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.2302158273381294,
|
| 456 |
+
"grad_norm": 0.6519774548706885,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.7192,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.2661870503597124,
|
| 463 |
+
"grad_norm": 0.7092939571259266,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.717,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.302158273381295,
|
| 470 |
+
"grad_norm": 0.8300683342338049,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.7171,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.338129496402878,
|
| 477 |
+
"grad_norm": 0.6364079517115279,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.7179,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.3741007194244603,
|
| 484 |
+
"grad_norm": 0.6830216482631195,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.7208,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.4100719424460433,
|
| 491 |
+
"grad_norm": 0.580810416113199,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.7201,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.446043165467626,
|
| 498 |
+
"grad_norm": 0.7709663647446697,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.7165,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.4820143884892087,
|
| 505 |
+
"grad_norm": 0.6587806242655105,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.7199,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.5179856115107913,
|
| 512 |
+
"grad_norm": 0.6679031168226195,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.7228,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.553956834532374,
|
| 519 |
+
"grad_norm": 0.5802019851320436,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.7211,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.5899280575539567,
|
| 526 |
+
"grad_norm": 0.633360775543426,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.7192,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.6258992805755397,
|
| 533 |
+
"grad_norm": 0.7014721250700231,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.7208,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.661870503597122,
|
| 540 |
+
"grad_norm": 0.5972726636881343,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.7184,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.697841726618705,
|
| 547 |
+
"grad_norm": 0.5454556975289979,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.7139,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.7338129496402876,
|
| 554 |
+
"grad_norm": 0.5626224999737693,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.7207,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.7697841726618706,
|
| 561 |
+
"grad_norm": 0.5106193565014756,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.7193,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.805755395683453,
|
| 568 |
+
"grad_norm": 0.6138738602878809,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.7185,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.841726618705036,
|
| 575 |
+
"grad_norm": 0.6093685279993987,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.7217,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.8776978417266186,
|
| 582 |
+
"grad_norm": 0.5564883285882788,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.7213,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.9136690647482015,
|
| 589 |
+
"grad_norm": 0.5906548449538034,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.7183,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.949640287769784,
|
| 596 |
+
"grad_norm": 0.5460219561244413,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.7216,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.985611510791367,
|
| 603 |
+
"grad_norm": 0.6453368774762195,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.7198,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 3.0,
|
| 610 |
+
"eval_loss": 0.7752296328544617,
|
| 611 |
+
"eval_runtime": 27.5746,
|
| 612 |
+
"eval_samples_per_second": 271.409,
|
| 613 |
+
"eval_steps_per_second": 1.088,
|
| 614 |
+
"step": 834
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.0,
|
| 618 |
+
"step": 834,
|
| 619 |
+
"total_flos": 1396981062696960.0,
|
| 620 |
+
"train_loss": 0.7675551453368555,
|
| 621 |
+
"train_runtime": 5571.5313,
|
| 622 |
+
"train_samples_per_second": 76.563,
|
| 623 |
+
"train_steps_per_second": 0.15
|
| 624 |
+
}
|
| 625 |
+
],
|
| 626 |
+
"logging_steps": 10,
|
| 627 |
+
"max_steps": 834,
|
| 628 |
+
"num_input_tokens_seen": 0,
|
| 629 |
+
"num_train_epochs": 3,
|
| 630 |
+
"save_steps": 500,
|
| 631 |
+
"stateful_callbacks": {
|
| 632 |
+
"TrainerControl": {
|
| 633 |
+
"args": {
|
| 634 |
+
"should_epoch_stop": false,
|
| 635 |
+
"should_evaluate": false,
|
| 636 |
+
"should_log": false,
|
| 637 |
+
"should_save": true,
|
| 638 |
+
"should_training_stop": true
|
| 639 |
+
},
|
| 640 |
+
"attributes": {}
|
| 641 |
+
}
|
| 642 |
+
},
|
| 643 |
+
"total_flos": 1396981062696960.0,
|
| 644 |
+
"train_batch_size": 16,
|
| 645 |
+
"trial_name": null,
|
| 646 |
+
"trial_params": null
|
| 647 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|