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 +511 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2-VL-2B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: models
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# models
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2-VL-2B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.0481
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2-VL-2B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: models
|
|
|
|
| 16 |
|
| 17 |
# models
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2-VL-2B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct) on the invoice_train dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.0481
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.04811210185289383,
|
| 4 |
+
"eval_runtime": 286.7587,
|
| 5 |
+
"eval_samples_per_second": 0.349,
|
| 6 |
+
"eval_steps_per_second": 0.349,
|
| 7 |
+
"total_flos": 3.3505112222539776e+16,
|
| 8 |
+
"train_loss": 0.005030520980556806,
|
| 9 |
+
"train_runtime": 5996.0343,
|
| 10 |
+
"train_samples_per_second": 0.4,
|
| 11 |
+
"train_steps_per_second": 0.1
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.04811210185289383,
|
| 4 |
+
"eval_runtime": 286.7587,
|
| 5 |
+
"eval_samples_per_second": 0.349,
|
| 6 |
+
"eval_steps_per_second": 0.349
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 3.3505112222539776e+16,
|
| 4 |
+
"train_loss": 0.005030520980556806,
|
| 5 |
+
"train_runtime": 5996.0343,
|
| 6 |
+
"train_samples_per_second": 0.4,
|
| 7 |
+
"train_steps_per_second": 0.1
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,511 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.05,
|
| 14 |
+
"grad_norm": 1.7140417098999023,
|
| 15 |
+
"learning_rate": 1.5e-05,
|
| 16 |
+
"loss": 0.7338,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.1,
|
| 21 |
+
"grad_norm": 1.1201739311218262,
|
| 22 |
+
"learning_rate": 3.1666666666666666e-05,
|
| 23 |
+
"loss": 0.3534,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.15,
|
| 28 |
+
"grad_norm": 1.3740745782852173,
|
| 29 |
+
"learning_rate": 4.8333333333333334e-05,
|
| 30 |
+
"loss": 0.1776,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2,
|
| 35 |
+
"grad_norm": 0.7445772886276245,
|
| 36 |
+
"learning_rate": 6.500000000000001e-05,
|
| 37 |
+
"loss": 0.1226,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.25,
|
| 42 |
+
"grad_norm": 0.6376151442527771,
|
| 43 |
+
"learning_rate": 8.166666666666667e-05,
|
| 44 |
+
"loss": 0.1086,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.3,
|
| 49 |
+
"grad_norm": 0.37051457166671753,
|
| 50 |
+
"learning_rate": 9.833333333333333e-05,
|
| 51 |
+
"loss": 0.0996,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.35,
|
| 56 |
+
"grad_norm": 0.5768002867698669,
|
| 57 |
+
"learning_rate": 9.99314767377287e-05,
|
| 58 |
+
"loss": 0.0959,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.4,
|
| 63 |
+
"grad_norm": 0.5846573114395142,
|
| 64 |
+
"learning_rate": 9.9694847320726e-05,
|
| 65 |
+
"loss": 0.0673,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.45,
|
| 70 |
+
"grad_norm": 0.5731233954429626,
|
| 71 |
+
"learning_rate": 9.929006627092299e-05,
|
| 72 |
+
"loss": 0.1235,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.5,
|
| 77 |
+
"grad_norm": 0.7275082468986511,
|
| 78 |
+
"learning_rate": 9.871850323926177e-05,
|
| 79 |
+
"loss": 0.0779,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.5,
|
| 84 |
+
"eval_loss": 0.0684906542301178,
|
| 85 |
+
"eval_runtime": 263.652,
|
| 86 |
+
"eval_samples_per_second": 0.379,
|
| 87 |
+
"eval_steps_per_second": 0.379,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.55,
|
| 92 |
+
"grad_norm": 0.5399947762489319,
|
| 93 |
+
"learning_rate": 9.798209221411747e-05,
|
| 94 |
+
"loss": 0.0591,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.6,
|
| 99 |
+
"grad_norm": 0.39991146326065063,
|
| 100 |
+
"learning_rate": 9.708332497729378e-05,
|
| 101 |
+
"loss": 0.0841,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.65,
|
| 106 |
+
"grad_norm": 0.4473568797111511,
|
| 107 |
+
"learning_rate": 9.602524267262203e-05,
|
| 108 |
+
"loss": 0.0823,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.7,
|
| 113 |
+
"grad_norm": 0.6712111234664917,
|
| 114 |
+
"learning_rate": 9.481142551569318e-05,
|
| 115 |
+
"loss": 0.0686,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.75,
|
| 120 |
+
"grad_norm": 0.3721703886985779,
|
| 121 |
+
"learning_rate": 9.344598067954152e-05,
|
| 122 |
+
"loss": 0.0658,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.8,
|
| 127 |
+
"grad_norm": 0.43187215924263,
|
| 128 |
+
"learning_rate": 9.193352839727121e-05,
|
| 129 |
+
"loss": 0.0669,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.85,
|
| 134 |
+
"grad_norm": 0.3821048140525818,
|
| 135 |
+
"learning_rate": 9.027918632864997e-05,
|
| 136 |
+
"loss": 0.0898,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.9,
|
| 141 |
+
"grad_norm": 0.7914325594902039,
|
| 142 |
+
"learning_rate": 8.848855224356839e-05,
|
| 143 |
+
"loss": 0.0777,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.95,
|
| 148 |
+
"grad_norm": 0.49602773785591125,
|
| 149 |
+
"learning_rate": 8.656768508095853e-05,
|
| 150 |
+
"loss": 0.0602,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.0,
|
| 155 |
+
"grad_norm": 0.36420896649360657,
|
| 156 |
+
"learning_rate": 8.452308444726249e-05,
|
| 157 |
+
"loss": 0.0647,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.0,
|
| 162 |
+
"eval_loss": 0.05108840763568878,
|
| 163 |
+
"eval_runtime": 263.56,
|
| 164 |
+
"eval_samples_per_second": 0.379,
|
| 165 |
+
"eval_steps_per_second": 0.379,
|
| 166 |
+
"step": 200
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.05,
|
| 170 |
+
"grad_norm": 0.29956021904945374,
|
| 171 |
+
"learning_rate": 8.236166862382163e-05,
|
| 172 |
+
"loss": 0.0381,
|
| 173 |
+
"step": 210
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.1,
|
| 177 |
+
"grad_norm": 0.3197619915008545,
|
| 178 |
+
"learning_rate": 8.009075115760243e-05,
|
| 179 |
+
"loss": 0.0478,
|
| 180 |
+
"step": 220
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.15,
|
| 184 |
+
"grad_norm": 0.43458399176597595,
|
| 185 |
+
"learning_rate": 7.771801611446858e-05,
|
| 186 |
+
"loss": 0.0349,
|
| 187 |
+
"step": 230
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.2,
|
| 191 |
+
"grad_norm": 0.3282864987850189,
|
| 192 |
+
"learning_rate": 7.52514920787345e-05,
|
| 193 |
+
"loss": 0.0309,
|
| 194 |
+
"step": 240
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.25,
|
| 198 |
+
"grad_norm": 0.24097760021686554,
|
| 199 |
+
"learning_rate": 7.269952498697734e-05,
|
| 200 |
+
"loss": 0.0209,
|
| 201 |
+
"step": 250
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.3,
|
| 205 |
+
"grad_norm": 0.31138163805007935,
|
| 206 |
+
"learning_rate": 7.007074988802946e-05,
|
| 207 |
+
"loss": 0.0414,
|
| 208 |
+
"step": 260
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.35,
|
| 212 |
+
"grad_norm": 0.2666899859905243,
|
| 213 |
+
"learning_rate": 6.737406172470657e-05,
|
| 214 |
+
"loss": 0.0592,
|
| 215 |
+
"step": 270
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.4,
|
| 219 |
+
"grad_norm": 0.513999879360199,
|
| 220 |
+
"learning_rate": 6.461858523613684e-05,
|
| 221 |
+
"loss": 0.0328,
|
| 222 |
+
"step": 280
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.45,
|
| 226 |
+
"grad_norm": 0.2508058547973633,
|
| 227 |
+
"learning_rate": 6.181364408253209e-05,
|
| 228 |
+
"loss": 0.0256,
|
| 229 |
+
"step": 290
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.5,
|
| 233 |
+
"grad_norm": 0.22155922651290894,
|
| 234 |
+
"learning_rate": 5.8968729296872874e-05,
|
| 235 |
+
"loss": 0.0292,
|
| 236 |
+
"step": 300
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.5,
|
| 240 |
+
"eval_loss": 0.04999532178044319,
|
| 241 |
+
"eval_runtime": 263.0996,
|
| 242 |
+
"eval_samples_per_second": 0.38,
|
| 243 |
+
"eval_steps_per_second": 0.38,
|
| 244 |
+
"step": 300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.55,
|
| 248 |
+
"grad_norm": 0.5395333170890808,
|
| 249 |
+
"learning_rate": 5.6093467170257374e-05,
|
| 250 |
+
"loss": 0.0468,
|
| 251 |
+
"step": 310
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.6,
|
| 255 |
+
"grad_norm": 0.28358444571495056,
|
| 256 |
+
"learning_rate": 5.319758667957928e-05,
|
| 257 |
+
"loss": 0.0418,
|
| 258 |
+
"step": 320
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.65,
|
| 262 |
+
"grad_norm": 0.18217401206493378,
|
| 263 |
+
"learning_rate": 5.0290886567749696e-05,
|
| 264 |
+
"loss": 0.0227,
|
| 265 |
+
"step": 330
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.7,
|
| 269 |
+
"grad_norm": 0.4007870852947235,
|
| 270 |
+
"learning_rate": 4.738320218785281e-05,
|
| 271 |
+
"loss": 0.0383,
|
| 272 |
+
"step": 340
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.75,
|
| 276 |
+
"grad_norm": 0.20039933919906616,
|
| 277 |
+
"learning_rate": 4.4484372223424415e-05,
|
| 278 |
+
"loss": 0.0371,
|
| 279 |
+
"step": 350
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.8,
|
| 283 |
+
"grad_norm": 0.12333797663450241,
|
| 284 |
+
"learning_rate": 4.160420539746115e-05,
|
| 285 |
+
"loss": 0.0331,
|
| 286 |
+
"step": 360
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.85,
|
| 290 |
+
"grad_norm": 0.3924310803413391,
|
| 291 |
+
"learning_rate": 3.875244728280676e-05,
|
| 292 |
+
"loss": 0.0396,
|
| 293 |
+
"step": 370
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.9,
|
| 297 |
+
"grad_norm": 0.07961348444223404,
|
| 298 |
+
"learning_rate": 3.593874732621847e-05,
|
| 299 |
+
"loss": 0.033,
|
| 300 |
+
"step": 380
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.95,
|
| 304 |
+
"grad_norm": 0.5689805746078491,
|
| 305 |
+
"learning_rate": 3.317262619769368e-05,
|
| 306 |
+
"loss": 0.0336,
|
| 307 |
+
"step": 390
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.0,
|
| 311 |
+
"grad_norm": 0.0888177827000618,
|
| 312 |
+
"learning_rate": 3.046344357553632e-05,
|
| 313 |
+
"loss": 0.028,
|
| 314 |
+
"step": 400
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.0,
|
| 318 |
+
"eval_loss": 0.04492698982357979,
|
| 319 |
+
"eval_runtime": 263.4857,
|
| 320 |
+
"eval_samples_per_second": 0.38,
|
| 321 |
+
"eval_steps_per_second": 0.38,
|
| 322 |
+
"step": 400
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.05,
|
| 326 |
+
"grad_norm": 0.4522015452384949,
|
| 327 |
+
"learning_rate": 2.7820366476168224e-05,
|
| 328 |
+
"loss": 0.0177,
|
| 329 |
+
"step": 410
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.1,
|
| 333 |
+
"grad_norm": 0.037934061139822006,
|
| 334 |
+
"learning_rate": 2.52523382358473e-05,
|
| 335 |
+
"loss": 0.0143,
|
| 336 |
+
"step": 420
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.15,
|
| 340 |
+
"grad_norm": 0.19100028276443481,
|
| 341 |
+
"learning_rate": 2.2768048249248648e-05,
|
| 342 |
+
"loss": 0.013,
|
| 343 |
+
"step": 430
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 2.2,
|
| 347 |
+
"grad_norm": 0.2139115184545517,
|
| 348 |
+
"learning_rate": 2.0375902567303472e-05,
|
| 349 |
+
"loss": 0.0105,
|
| 350 |
+
"step": 440
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.25,
|
| 354 |
+
"grad_norm": 0.31156909465789795,
|
| 355 |
+
"learning_rate": 1.80839954537836e-05,
|
| 356 |
+
"loss": 0.0136,
|
| 357 |
+
"step": 450
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.3,
|
| 361 |
+
"grad_norm": 0.08543579280376434,
|
| 362 |
+
"learning_rate": 1.5900081996875083e-05,
|
| 363 |
+
"loss": 0.0196,
|
| 364 |
+
"step": 460
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 2.35,
|
| 368 |
+
"grad_norm": 0.10310215502977371,
|
| 369 |
+
"learning_rate": 1.3831551868414599e-05,
|
| 370 |
+
"loss": 0.017,
|
| 371 |
+
"step": 470
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.4,
|
| 375 |
+
"grad_norm": 0.2742729187011719,
|
| 376 |
+
"learning_rate": 1.1885404319579108e-05,
|
| 377 |
+
"loss": 0.0254,
|
| 378 |
+
"step": 480
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.45,
|
| 382 |
+
"grad_norm": 0.09017336368560791,
|
| 383 |
+
"learning_rate": 1.006822449763537e-05,
|
| 384 |
+
"loss": 0.0174,
|
| 385 |
+
"step": 490
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.5,
|
| 389 |
+
"grad_norm": 0.07442392408847809,
|
| 390 |
+
"learning_rate": 8.38616116388612e-06,
|
| 391 |
+
"loss": 0.013,
|
| 392 |
+
"step": 500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.5,
|
| 396 |
+
"eval_loss": 0.048839278519153595,
|
| 397 |
+
"eval_runtime": 284.8522,
|
| 398 |
+
"eval_samples_per_second": 0.351,
|
| 399 |
+
"eval_steps_per_second": 0.351,
|
| 400 |
+
"step": 500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.55,
|
| 404 |
+
"grad_norm": 0.29669731855392456,
|
| 405 |
+
"learning_rate": 6.844905888208181e-06,
|
| 406 |
+
"loss": 0.015,
|
| 407 |
+
"step": 510
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.6,
|
| 411 |
+
"grad_norm": 0.3099665939807892,
|
| 412 |
+
"learning_rate": 5.449673790581611e-06,
|
| 413 |
+
"loss": 0.0148,
|
| 414 |
+
"step": 520
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.65,
|
| 418 |
+
"grad_norm": 0.20635627210140228,
|
| 419 |
+
"learning_rate": 4.205185894774455e-06,
|
| 420 |
+
"loss": 0.0156,
|
| 421 |
+
"step": 530
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.7,
|
| 425 |
+
"grad_norm": 0.2796818017959595,
|
| 426 |
+
"learning_rate": 3.115653153892761e-06,
|
| 427 |
+
"loss": 0.015,
|
| 428 |
+
"step": 540
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.75,
|
| 432 |
+
"grad_norm": 0.15870630741119385,
|
| 433 |
+
"learning_rate": 2.1847622018482283e-06,
|
| 434 |
+
"loss": 0.0144,
|
| 435 |
+
"step": 550
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.8,
|
| 439 |
+
"grad_norm": 0.23704785108566284,
|
| 440 |
+
"learning_rate": 1.4156628789559922e-06,
|
| 441 |
+
"loss": 0.0241,
|
| 442 |
+
"step": 560
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 2.85,
|
| 446 |
+
"grad_norm": 0.025902193039655685,
|
| 447 |
+
"learning_rate": 8.10957573872062e-07,
|
| 448 |
+
"loss": 0.0096,
|
| 449 |
+
"step": 570
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.9,
|
| 453 |
+
"grad_norm": 0.04191463440656662,
|
| 454 |
+
"learning_rate": 3.7269241793390085e-07,
|
| 455 |
+
"loss": 0.0086,
|
| 456 |
+
"step": 580
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.95,
|
| 460 |
+
"grad_norm": 0.04461716488003731,
|
| 461 |
+
"learning_rate": 1.0235036169963242e-07,
|
| 462 |
+
"loss": 0.0115,
|
| 463 |
+
"step": 590
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.0,
|
| 467 |
+
"grad_norm": 0.21583612263202667,
|
| 468 |
+
"learning_rate": 8.461571127882373e-10,
|
| 469 |
+
"loss": 0.0116,
|
| 470 |
+
"step": 600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.0,
|
| 474 |
+
"eval_loss": 0.04811210185289383,
|
| 475 |
+
"eval_runtime": 284.4632,
|
| 476 |
+
"eval_samples_per_second": 0.352,
|
| 477 |
+
"eval_steps_per_second": 0.352,
|
| 478 |
+
"step": 600
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 3.0,
|
| 482 |
+
"step": 600,
|
| 483 |
+
"total_flos": 3.3505112222539776e+16,
|
| 484 |
+
"train_loss": 0.005030520980556806,
|
| 485 |
+
"train_runtime": 5996.0343,
|
| 486 |
+
"train_samples_per_second": 0.4,
|
| 487 |
+
"train_steps_per_second": 0.1
|
| 488 |
+
}
|
| 489 |
+
],
|
| 490 |
+
"logging_steps": 10,
|
| 491 |
+
"max_steps": 600,
|
| 492 |
+
"num_input_tokens_seen": 0,
|
| 493 |
+
"num_train_epochs": 3,
|
| 494 |
+
"save_steps": 200,
|
| 495 |
+
"stateful_callbacks": {
|
| 496 |
+
"TrainerControl": {
|
| 497 |
+
"args": {
|
| 498 |
+
"should_epoch_stop": false,
|
| 499 |
+
"should_evaluate": false,
|
| 500 |
+
"should_log": false,
|
| 501 |
+
"should_save": true,
|
| 502 |
+
"should_training_stop": true
|
| 503 |
+
},
|
| 504 |
+
"attributes": {}
|
| 505 |
+
}
|
| 506 |
+
},
|
| 507 |
+
"total_flos": 3.3505112222539776e+16,
|
| 508 |
+
"train_batch_size": 1,
|
| 509 |
+
"trial_name": null,
|
| 510 |
+
"trial_params": null
|
| 511 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|