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: oh_v1.3_alpaca_x2
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_v1.3_alpaca_x2
|
| 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: 0.7331
|
| 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_alpaca_x2
|
|
|
|
| 16 |
|
| 17 |
# oh_v1.3_alpaca_x2
|
| 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_alpaca_x2 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7331
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.995973154362416,
|
| 3 |
+
"eval_loss": 0.733113169670105,
|
| 4 |
+
"eval_runtime": 297.3971,
|
| 5 |
+
"eval_samples_per_second": 25.313,
|
| 6 |
+
"eval_steps_per_second": 0.397,
|
| 7 |
+
"total_flos": 1401796794777600.0,
|
| 8 |
+
"train_loss": 0.7062144683965503,
|
| 9 |
+
"train_runtime": 49420.9289,
|
| 10 |
+
"train_samples_per_second": 8.682,
|
| 11 |
+
"train_steps_per_second": 0.017
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.995973154362416,
|
| 3 |
+
"eval_loss": 0.733113169670105,
|
| 4 |
+
"eval_runtime": 297.3971,
|
| 5 |
+
"eval_samples_per_second": 25.313,
|
| 6 |
+
"eval_steps_per_second": 0.397
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.995973154362416,
|
| 3 |
+
"total_flos": 1401796794777600.0,
|
| 4 |
+
"train_loss": 0.7062144683965503,
|
| 5 |
+
"train_runtime": 49420.9289,
|
| 6 |
+
"train_samples_per_second": 8.682,
|
| 7 |
+
"train_steps_per_second": 0.017
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,647 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.995973154362416,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 837,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.035794183445190156,
|
| 13 |
+
"grad_norm": 8.401638886940637,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.0542,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07158836689038031,
|
| 20 |
+
"grad_norm": 2.3991823729158823,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.9185,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10738255033557047,
|
| 27 |
+
"grad_norm": 1.8009395703582636,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.8832,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14317673378076062,
|
| 34 |
+
"grad_norm": 1.1993029426540307,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.8514,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1789709172259508,
|
| 41 |
+
"grad_norm": 1.2429515479763662,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.8348,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.21476510067114093,
|
| 48 |
+
"grad_norm": 1.2280588442944276,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.8123,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2505592841163311,
|
| 55 |
+
"grad_norm": 1.0812304059733735,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.803,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.28635346756152125,
|
| 62 |
+
"grad_norm": 1.0179970959347673,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.7897,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3221476510067114,
|
| 69 |
+
"grad_norm": 1.3275681971077118,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.7885,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3579418344519016,
|
| 76 |
+
"grad_norm": 0.9714888938151465,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.7772,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.39373601789709173,
|
| 83 |
+
"grad_norm": 1.0884614950165932,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.7786,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.42953020134228187,
|
| 90 |
+
"grad_norm": 1.45656642485753,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.7684,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.465324384787472,
|
| 97 |
+
"grad_norm": 0.7010636928500263,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.7688,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5011185682326622,
|
| 104 |
+
"grad_norm": 0.7397488137262136,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.7706,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5369127516778524,
|
| 111 |
+
"grad_norm": 0.9526425688427698,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.762,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5727069351230425,
|
| 118 |
+
"grad_norm": 0.7088169136590493,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.7639,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6085011185682326,
|
| 125 |
+
"grad_norm": 0.7571173656145734,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.7568,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6442953020134228,
|
| 132 |
+
"grad_norm": 0.6372701282417548,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.7593,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.680089485458613,
|
| 139 |
+
"grad_norm": 0.5399460206729987,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.752,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.7158836689038032,
|
| 146 |
+
"grad_norm": 0.7186356757703609,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.7535,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.7516778523489933,
|
| 153 |
+
"grad_norm": 0.6839088392647745,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.7509,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7874720357941835,
|
| 160 |
+
"grad_norm": 0.587117040709812,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.7572,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.8232662192393736,
|
| 167 |
+
"grad_norm": 0.679563598992802,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.7445,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8590604026845637,
|
| 174 |
+
"grad_norm": 0.5683321619095465,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.7474,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8948545861297539,
|
| 181 |
+
"grad_norm": 0.71166875890661,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.7497,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.930648769574944,
|
| 188 |
+
"grad_norm": 0.8045131045063768,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.7463,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9664429530201343,
|
| 195 |
+
"grad_norm": 0.7352070256803627,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.7406,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9986577181208054,
|
| 202 |
+
"eval_loss": 0.7448742389678955,
|
| 203 |
+
"eval_runtime": 293.8562,
|
| 204 |
+
"eval_samples_per_second": 25.618,
|
| 205 |
+
"eval_steps_per_second": 0.402,
|
| 206 |
+
"step": 279
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.0022371364653244,
|
| 210 |
+
"grad_norm": 1.1521591695523132,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.783,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.0380313199105144,
|
| 217 |
+
"grad_norm": 0.9493139325679057,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.6946,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.0738255033557047,
|
| 224 |
+
"grad_norm": 1.2157290004753143,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.6923,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.109619686800895,
|
| 231 |
+
"grad_norm": 1.3499555504652312,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.6925,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.145413870246085,
|
| 238 |
+
"grad_norm": 1.245428077765806,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.6938,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.1812080536912752,
|
| 245 |
+
"grad_norm": 1.0729932571470133,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.6897,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.2170022371364653,
|
| 252 |
+
"grad_norm": 0.6864034681556961,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6971,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.2527964205816555,
|
| 259 |
+
"grad_norm": 0.6794551859864176,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.6912,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.2885906040268456,
|
| 266 |
+
"grad_norm": 0.7163040868054562,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.6892,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.3243847874720358,
|
| 273 |
+
"grad_norm": 0.6048987435825018,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.6895,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.360178970917226,
|
| 280 |
+
"grad_norm": 0.5792902824967668,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.6887,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.395973154362416,
|
| 287 |
+
"grad_norm": 0.7880828049853106,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.6868,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.4317673378076063,
|
| 294 |
+
"grad_norm": 0.7894027755224914,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.6862,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.4675615212527964,
|
| 301 |
+
"grad_norm": 0.8577775477955841,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.6887,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.5033557046979866,
|
| 308 |
+
"grad_norm": 0.6168297625737176,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.6866,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.5391498881431769,
|
| 315 |
+
"grad_norm": 0.7366230838210155,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.6904,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.574944071588367,
|
| 322 |
+
"grad_norm": 0.6622652592180838,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.6944,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.610738255033557,
|
| 329 |
+
"grad_norm": 0.569716627527462,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.6848,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.6465324384787472,
|
| 336 |
+
"grad_norm": 0.6235847507038318,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.6873,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.6823266219239374,
|
| 343 |
+
"grad_norm": 0.6482653709805161,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.6834,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.7181208053691275,
|
| 350 |
+
"grad_norm": 0.5969794316920956,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.6831,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.7539149888143175,
|
| 357 |
+
"grad_norm": 0.6436900491750924,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.6842,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.7897091722595078,
|
| 364 |
+
"grad_norm": 0.5633091053168701,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.687,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.825503355704698,
|
| 371 |
+
"grad_norm": 0.6798124363156657,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.6891,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.8612975391498883,
|
| 378 |
+
"grad_norm": 0.6792397317268221,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.6906,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.8970917225950783,
|
| 385 |
+
"grad_norm": 0.6285135955779139,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.684,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.9328859060402683,
|
| 392 |
+
"grad_norm": 0.5601176877036173,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.6882,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.9686800894854586,
|
| 399 |
+
"grad_norm": 0.7243308708474905,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.6867,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.9973154362416108,
|
| 406 |
+
"eval_loss": 0.7317517399787903,
|
| 407 |
+
"eval_runtime": 294.6356,
|
| 408 |
+
"eval_samples_per_second": 25.55,
|
| 409 |
+
"eval_steps_per_second": 0.4,
|
| 410 |
+
"step": 558
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.004474272930649,
|
| 414 |
+
"grad_norm": 0.99037695759716,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.7203,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.040268456375839,
|
| 421 |
+
"grad_norm": 0.9902193880849315,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.6303,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.076062639821029,
|
| 428 |
+
"grad_norm": 0.6397622212118742,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.6299,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.111856823266219,
|
| 435 |
+
"grad_norm": 0.7917691287759923,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.6341,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.1476510067114094,
|
| 442 |
+
"grad_norm": 0.6240987698183383,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.6354,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.1834451901565997,
|
| 449 |
+
"grad_norm": 0.7599713166789355,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.6307,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.21923937360179,
|
| 456 |
+
"grad_norm": 0.7131126040530179,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.6333,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.2550335570469797,
|
| 463 |
+
"grad_norm": 0.8259088835615077,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.6358,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.29082774049217,
|
| 470 |
+
"grad_norm": 0.6541826430581148,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.6352,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.3266219239373602,
|
| 477 |
+
"grad_norm": 0.7050135787324793,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.6355,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.3624161073825505,
|
| 484 |
+
"grad_norm": 0.7052621683849559,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.6337,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.3982102908277403,
|
| 491 |
+
"grad_norm": 0.7256733119248161,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.6361,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.4340044742729305,
|
| 498 |
+
"grad_norm": 0.7161998050929107,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.6363,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.469798657718121,
|
| 505 |
+
"grad_norm": 0.6120627733035356,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.6361,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.505592841163311,
|
| 512 |
+
"grad_norm": 0.6865307447351127,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.6357,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.5413870246085013,
|
| 519 |
+
"grad_norm": 0.6728771691791204,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.6378,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.577181208053691,
|
| 526 |
+
"grad_norm": 1.1082380037070934,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.6364,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.6129753914988814,
|
| 533 |
+
"grad_norm": 0.6069652328826274,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.6397,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.6487695749440716,
|
| 540 |
+
"grad_norm": 0.7205831497884968,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.6369,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.684563758389262,
|
| 547 |
+
"grad_norm": 0.6012406001171214,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.6382,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.720357941834452,
|
| 554 |
+
"grad_norm": 0.5859459819180776,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.6353,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.756152125279642,
|
| 561 |
+
"grad_norm": 0.7431570789548247,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.6398,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.791946308724832,
|
| 568 |
+
"grad_norm": 0.6535586087228809,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.6411,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.8277404921700224,
|
| 575 |
+
"grad_norm": 0.6893473487499976,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.6367,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.8635346756152127,
|
| 582 |
+
"grad_norm": 0.6911347153406258,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.6361,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.899328859060403,
|
| 589 |
+
"grad_norm": 0.8336045673864524,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.6442,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.9351230425055927,
|
| 596 |
+
"grad_norm": 0.730089605426641,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.6395,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.970917225950783,
|
| 603 |
+
"grad_norm": 0.68005844887895,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.64,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.995973154362416,
|
| 610 |
+
"eval_loss": 0.733113169670105,
|
| 611 |
+
"eval_runtime": 295.0877,
|
| 612 |
+
"eval_samples_per_second": 25.511,
|
| 613 |
+
"eval_steps_per_second": 0.4,
|
| 614 |
+
"step": 837
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.995973154362416,
|
| 618 |
+
"step": 837,
|
| 619 |
+
"total_flos": 1401796794777600.0,
|
| 620 |
+
"train_loss": 0.7062144683965503,
|
| 621 |
+
"train_runtime": 49420.9289,
|
| 622 |
+
"train_samples_per_second": 8.682,
|
| 623 |
+
"train_steps_per_second": 0.017
|
| 624 |
+
}
|
| 625 |
+
],
|
| 626 |
+
"logging_steps": 10,
|
| 627 |
+
"max_steps": 837,
|
| 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": 1401796794777600.0,
|
| 644 |
+
"train_batch_size": 8,
|
| 645 |
+
"trial_name": null,
|
| 646 |
+
"trial_params": null
|
| 647 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|