Commit
•
3850af7
1
Parent(s):
08a7046
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 +423 -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/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_qwen_bsz1024_dcftv1.2
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_qwen_bsz1024_dcftv1.2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.6337
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_qwen_bsz1024_dcftv1.2
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_qwen_bsz1024_dcftv1.2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the mlfoundations-dev/oh-dcft-v1.2_no-curation_gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6337
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.990490124359912,
|
| 3 |
+
"eval_loss": 0.63369220495224,
|
| 4 |
+
"eval_runtime": 171.6506,
|
| 5 |
+
"eval_samples_per_second": 53.65,
|
| 6 |
+
"eval_steps_per_second": 0.419,
|
| 7 |
+
"total_flos": 2138433883471872.0,
|
| 8 |
+
"train_loss": 0.6292863135244332,
|
| 9 |
+
"train_runtime": 27713.9731,
|
| 10 |
+
"train_samples_per_second": 18.939,
|
| 11 |
+
"train_steps_per_second": 0.018
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.990490124359912,
|
| 3 |
+
"eval_loss": 0.63369220495224,
|
| 4 |
+
"eval_runtime": 171.6506,
|
| 5 |
+
"eval_samples_per_second": 53.65,
|
| 6 |
+
"eval_steps_per_second": 0.419
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.990490124359912,
|
| 3 |
+
"total_flos": 2138433883471872.0,
|
| 4 |
+
"train_loss": 0.6292863135244332,
|
| 5 |
+
"train_runtime": 27713.9731,
|
| 6 |
+
"train_samples_per_second": 18.939,
|
| 7 |
+
"train_steps_per_second": 0.018
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,423 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.990490124359912,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 510,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.058522311631309436,
|
| 13 |
+
"grad_norm": 1.5647390529813434,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.8022,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11704462326261887,
|
| 20 |
+
"grad_norm": 2.4298906740936874,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7306,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1755669348939283,
|
| 27 |
+
"grad_norm": 1.1278550883615037,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7122,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.23408924652523774,
|
| 34 |
+
"grad_norm": 1.1246574687423805,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6975,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.29261155815654716,
|
| 41 |
+
"grad_norm": 1.0811775432928663,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6821,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3511338697878566,
|
| 48 |
+
"grad_norm": 0.8345121386846462,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6822,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.40965618141916604,
|
| 55 |
+
"grad_norm": 0.502423577542533,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6631,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4681784930504755,
|
| 62 |
+
"grad_norm": 0.3206403744702351,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6566,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.5267008046817849,
|
| 69 |
+
"grad_norm": 0.3373586439028653,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6613,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.5852231163130943,
|
| 76 |
+
"grad_norm": 0.27440465078078524,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6497,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.6437454279444038,
|
| 83 |
+
"grad_norm": 0.25729298157504654,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6506,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7022677395757132,
|
| 90 |
+
"grad_norm": 0.2774645357576214,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6479,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.7607900512070227,
|
| 97 |
+
"grad_norm": 0.2750786001903561,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6509,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8193123628383321,
|
| 104 |
+
"grad_norm": 0.3018000353503424,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6532,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.8778346744696416,
|
| 111 |
+
"grad_norm": 0.2651836266343764,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6407,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.936356986100951,
|
| 118 |
+
"grad_norm": 0.2621590800809169,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.646,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.9948792977322605,
|
| 125 |
+
"grad_norm": 0.274425449696112,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6429,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.9948792977322605,
|
| 132 |
+
"eval_loss": 0.6455708742141724,
|
| 133 |
+
"eval_runtime": 172.3102,
|
| 134 |
+
"eval_samples_per_second": 53.444,
|
| 135 |
+
"eval_steps_per_second": 0.418,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.0563277249451353,
|
| 140 |
+
"grad_norm": 0.32659804954217203,
|
| 141 |
+
"learning_rate": 5e-06,
|
| 142 |
+
"loss": 0.6596,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.1148500365764447,
|
| 147 |
+
"grad_norm": 0.29235225467356285,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.6219,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.1733723482077543,
|
| 154 |
+
"grad_norm": 0.2686911846272285,
|
| 155 |
+
"learning_rate": 5e-06,
|
| 156 |
+
"loss": 0.6246,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.2318946598390637,
|
| 161 |
+
"grad_norm": 0.2689856133611371,
|
| 162 |
+
"learning_rate": 5e-06,
|
| 163 |
+
"loss": 0.618,
|
| 164 |
+
"step": 210
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.290416971470373,
|
| 168 |
+
"grad_norm": 0.26872283242131406,
|
| 169 |
+
"learning_rate": 5e-06,
|
| 170 |
+
"loss": 0.6202,
|
| 171 |
+
"step": 220
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.3489392831016827,
|
| 175 |
+
"grad_norm": 0.301091252809549,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.618,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.4074615947329918,
|
| 182 |
+
"grad_norm": 0.2920775430394786,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.6142,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4659839063643014,
|
| 189 |
+
"grad_norm": 0.2456820075171799,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.6155,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.5245062179956108,
|
| 196 |
+
"grad_norm": 0.2938378044663654,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.6187,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.5830285296269202,
|
| 203 |
+
"grad_norm": 0.32438651891226156,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.6219,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.6415508412582298,
|
| 210 |
+
"grad_norm": 0.25545801371272864,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.616,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.700073152889539,
|
| 217 |
+
"grad_norm": 0.26294073057220163,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.6127,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.7585954645208486,
|
| 224 |
+
"grad_norm": 0.26462245389002803,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.6168,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.817117776152158,
|
| 231 |
+
"grad_norm": 0.2847262707293318,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.6172,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.8756400877834674,
|
| 238 |
+
"grad_norm": 0.2669714428041422,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.614,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.934162399414777,
|
| 245 |
+
"grad_norm": 0.25457144598514,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.6166,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.9926847110460864,
|
| 252 |
+
"grad_norm": 0.2608967910015083,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6126,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.9926847110460864,
|
| 259 |
+
"eval_loss": 0.6363422274589539,
|
| 260 |
+
"eval_runtime": 172.306,
|
| 261 |
+
"eval_samples_per_second": 53.446,
|
| 262 |
+
"eval_steps_per_second": 0.418,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.0541331382589614,
|
| 267 |
+
"grad_norm": 0.271883921683299,
|
| 268 |
+
"learning_rate": 5e-06,
|
| 269 |
+
"loss": 0.6297,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.1126554498902705,
|
| 274 |
+
"grad_norm": 0.24729272080119263,
|
| 275 |
+
"learning_rate": 5e-06,
|
| 276 |
+
"loss": 0.5897,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.17117776152158,
|
| 281 |
+
"grad_norm": 0.27092891797600144,
|
| 282 |
+
"learning_rate": 5e-06,
|
| 283 |
+
"loss": 0.5946,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.2297000731528893,
|
| 288 |
+
"grad_norm": 0.3032127102208398,
|
| 289 |
+
"learning_rate": 5e-06,
|
| 290 |
+
"loss": 0.594,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.288222384784199,
|
| 295 |
+
"grad_norm": 0.25853126440367846,
|
| 296 |
+
"learning_rate": 5e-06,
|
| 297 |
+
"loss": 0.588,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.3467446964155085,
|
| 302 |
+
"grad_norm": 0.3077689025344159,
|
| 303 |
+
"learning_rate": 5e-06,
|
| 304 |
+
"loss": 0.5943,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.4052670080468177,
|
| 309 |
+
"grad_norm": 0.2827487146132787,
|
| 310 |
+
"learning_rate": 5e-06,
|
| 311 |
+
"loss": 0.5933,
|
| 312 |
+
"step": 410
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.4637893196781273,
|
| 316 |
+
"grad_norm": 0.2519214403191199,
|
| 317 |
+
"learning_rate": 5e-06,
|
| 318 |
+
"loss": 0.5898,
|
| 319 |
+
"step": 420
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.522311631309437,
|
| 323 |
+
"grad_norm": 0.2751668540595721,
|
| 324 |
+
"learning_rate": 5e-06,
|
| 325 |
+
"loss": 0.588,
|
| 326 |
+
"step": 430
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.580833942940746,
|
| 330 |
+
"grad_norm": 0.2530698336402752,
|
| 331 |
+
"learning_rate": 5e-06,
|
| 332 |
+
"loss": 0.5883,
|
| 333 |
+
"step": 440
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.6393562545720557,
|
| 337 |
+
"grad_norm": 0.25471213766207895,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.5951,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.6978785662033653,
|
| 344 |
+
"grad_norm": 0.29077003251470107,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.5914,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.7564008778346745,
|
| 351 |
+
"grad_norm": 0.30152118910674564,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.5917,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.8149231894659836,
|
| 358 |
+
"grad_norm": 0.26709177034419973,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.5923,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.8734455010972932,
|
| 365 |
+
"grad_norm": 0.23395614388611538,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.5888,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.931967812728603,
|
| 372 |
+
"grad_norm": 0.28669880402317394,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.5865,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.990490124359912,
|
| 379 |
+
"grad_norm": 0.2531977873715163,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.5898,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.990490124359912,
|
| 386 |
+
"eval_loss": 0.63369220495224,
|
| 387 |
+
"eval_runtime": 171.9297,
|
| 388 |
+
"eval_samples_per_second": 53.563,
|
| 389 |
+
"eval_steps_per_second": 0.419,
|
| 390 |
+
"step": 510
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 2.990490124359912,
|
| 394 |
+
"step": 510,
|
| 395 |
+
"total_flos": 2138433883471872.0,
|
| 396 |
+
"train_loss": 0.6292863135244332,
|
| 397 |
+
"train_runtime": 27713.9731,
|
| 398 |
+
"train_samples_per_second": 18.939,
|
| 399 |
+
"train_steps_per_second": 0.018
|
| 400 |
+
}
|
| 401 |
+
],
|
| 402 |
+
"logging_steps": 10,
|
| 403 |
+
"max_steps": 510,
|
| 404 |
+
"num_input_tokens_seen": 0,
|
| 405 |
+
"num_train_epochs": 3,
|
| 406 |
+
"save_steps": 500,
|
| 407 |
+
"stateful_callbacks": {
|
| 408 |
+
"TrainerControl": {
|
| 409 |
+
"args": {
|
| 410 |
+
"should_epoch_stop": false,
|
| 411 |
+
"should_evaluate": false,
|
| 412 |
+
"should_log": false,
|
| 413 |
+
"should_save": true,
|
| 414 |
+
"should_training_stop": true
|
| 415 |
+
},
|
| 416 |
+
"attributes": {}
|
| 417 |
+
}
|
| 418 |
+
},
|
| 419 |
+
"total_flos": 2138433883471872.0,
|
| 420 |
+
"train_batch_size": 8,
|
| 421 |
+
"trial_name": null,
|
| 422 |
+
"trial_params": null
|
| 423 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|