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 +528 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: gemma
|
|
| 4 |
base_model: google/gemma-2-9b
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_gemma_bsz1024
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_gemma_bsz1024
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5917
|
| 21 |
|
|
|
|
| 4 |
base_model: google/gemma-2-9b
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_gemma_bsz1024
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_gemma_bsz1024
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the mlfoundations-dev/oh-dcft-v3.1-gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5917
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9946403385049365,
|
| 3 |
+
"eval_loss": 0.5917297005653381,
|
| 4 |
+
"eval_runtime": 356.383,
|
| 5 |
+
"eval_samples_per_second": 33.506,
|
| 6 |
+
"eval_steps_per_second": 0.525,
|
| 7 |
+
"total_flos": 5052524767739904.0,
|
| 8 |
+
"train_loss": 0.5641303292586612,
|
| 9 |
+
"train_runtime": 61418.5213,
|
| 10 |
+
"train_samples_per_second": 11.081,
|
| 11 |
+
"train_steps_per_second": 0.011
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9946403385049365,
|
| 3 |
+
"eval_loss": 0.5917297005653381,
|
| 4 |
+
"eval_runtime": 356.383,
|
| 5 |
+
"eval_samples_per_second": 33.506,
|
| 6 |
+
"eval_steps_per_second": 0.525
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9946403385049365,
|
| 3 |
+
"total_flos": 5052524767739904.0,
|
| 4 |
+
"train_loss": 0.5641303292586612,
|
| 5 |
+
"train_runtime": 61418.5213,
|
| 6 |
+
"train_samples_per_second": 11.081,
|
| 7 |
+
"train_steps_per_second": 0.011
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9946403385049365,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 663,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.045133991537376586,
|
| 13 |
+
"grad_norm": 1.0502111381705257,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7438,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09026798307475317,
|
| 20 |
+
"grad_norm": 0.7055928208506984,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6709,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.13540197461212977,
|
| 27 |
+
"grad_norm": 0.5033134209699918,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6463,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.18053596614950634,
|
| 34 |
+
"grad_norm": 0.618156041678536,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6389,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.22566995768688294,
|
| 41 |
+
"grad_norm": 0.49695878808904403,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6349,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.27080394922425954,
|
| 48 |
+
"grad_norm": 0.5844890023067665,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6247,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.3159379407616361,
|
| 55 |
+
"grad_norm": 0.6341705695305015,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6227,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3610719322990127,
|
| 62 |
+
"grad_norm": 0.9851855606862004,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6225,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.40620592383638926,
|
| 69 |
+
"grad_norm": 0.6200620013182269,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6161,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.4513399153737659,
|
| 76 |
+
"grad_norm": 0.4866582707983752,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6118,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.49647390691114246,
|
| 83 |
+
"grad_norm": 0.4846006073335934,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6071,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.5416078984485191,
|
| 90 |
+
"grad_norm": 0.4640299470151304,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6108,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5867418899858956,
|
| 97 |
+
"grad_norm": 0.5199888815602103,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6056,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.6318758815232722,
|
| 104 |
+
"grad_norm": 0.5702031853447341,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6056,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6770098730606487,
|
| 111 |
+
"grad_norm": 0.8001760897422481,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6054,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7221438645980254,
|
| 118 |
+
"grad_norm": 0.5587666551476193,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6031,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.767277856135402,
|
| 125 |
+
"grad_norm": 0.43674759924584844,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.5996,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8124118476727785,
|
| 132 |
+
"grad_norm": 0.5559456367161821,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6003,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.8575458392101551,
|
| 139 |
+
"grad_norm": 0.44306696302984344,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6049,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9026798307475318,
|
| 146 |
+
"grad_norm": 0.5092942073450014,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.5985,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9478138222849083,
|
| 153 |
+
"grad_norm": 0.47841352538603515,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.5975,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9929478138222849,
|
| 160 |
+
"grad_norm": 0.757057117339616,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.5944,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.9974612129760225,
|
| 167 |
+
"eval_loss": 0.5952667593955994,
|
| 168 |
+
"eval_runtime": 354.3828,
|
| 169 |
+
"eval_samples_per_second": 33.695,
|
| 170 |
+
"eval_steps_per_second": 0.528,
|
| 171 |
+
"step": 221
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.039210155148096,
|
| 175 |
+
"grad_norm": 0.6288422422717695,
|
| 176 |
+
"learning_rate": 5e-06,
|
| 177 |
+
"loss": 0.5957,
|
| 178 |
+
"step": 230
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.0843441466854724,
|
| 182 |
+
"grad_norm": 0.5368840934195099,
|
| 183 |
+
"learning_rate": 5e-06,
|
| 184 |
+
"loss": 0.5546,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.1294781382228491,
|
| 189 |
+
"grad_norm": 0.5562178733269311,
|
| 190 |
+
"learning_rate": 5e-06,
|
| 191 |
+
"loss": 0.5576,
|
| 192 |
+
"step": 250
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.1746121297602257,
|
| 196 |
+
"grad_norm": 0.44566707316498066,
|
| 197 |
+
"learning_rate": 5e-06,
|
| 198 |
+
"loss": 0.5531,
|
| 199 |
+
"step": 260
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.2197461212976022,
|
| 203 |
+
"grad_norm": 0.5774903613750582,
|
| 204 |
+
"learning_rate": 5e-06,
|
| 205 |
+
"loss": 0.5549,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.264880112834979,
|
| 210 |
+
"grad_norm": 0.429832871209753,
|
| 211 |
+
"learning_rate": 5e-06,
|
| 212 |
+
"loss": 0.5572,
|
| 213 |
+
"step": 280
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.3100141043723554,
|
| 217 |
+
"grad_norm": 0.5244854621545229,
|
| 218 |
+
"learning_rate": 5e-06,
|
| 219 |
+
"loss": 0.5565,
|
| 220 |
+
"step": 290
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.355148095909732,
|
| 224 |
+
"grad_norm": 0.50009636557672,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.5587,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.4002820874471085,
|
| 231 |
+
"grad_norm": 0.46468247492051845,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5557,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4454160789844852,
|
| 238 |
+
"grad_norm": 0.6185616791546158,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.555,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4905500705218617,
|
| 245 |
+
"grad_norm": 0.5522113108694092,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5578,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.5356840620592385,
|
| 252 |
+
"grad_norm": 0.5687883176073543,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.5519,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.580818053596615,
|
| 259 |
+
"grad_norm": 0.4317341125326038,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5523,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.6259520451339915,
|
| 266 |
+
"grad_norm": 0.4415467428367944,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.561,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.671086036671368,
|
| 273 |
+
"grad_norm": 0.5437873630019581,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.5562,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.7162200282087448,
|
| 280 |
+
"grad_norm": 0.46394331744324036,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5552,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.7613540197461213,
|
| 287 |
+
"grad_norm": 0.4475505593561043,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5528,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.806488011283498,
|
| 294 |
+
"grad_norm": 0.5295023528850353,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5517,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.8516220028208745,
|
| 301 |
+
"grad_norm": 0.5025367324019494,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5543,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.896755994358251,
|
| 308 |
+
"grad_norm": 0.46331161327776976,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.5578,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.9418899858956276,
|
| 315 |
+
"grad_norm": 0.5483627384227747,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5486,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.987023977433004,
|
| 322 |
+
"grad_norm": 0.4673744718040457,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5591,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.9960507757404795,
|
| 329 |
+
"eval_loss": 0.5866958498954773,
|
| 330 |
+
"eval_runtime": 356.554,
|
| 331 |
+
"eval_samples_per_second": 33.49,
|
| 332 |
+
"eval_steps_per_second": 0.524,
|
| 333 |
+
"step": 442
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.0332863187588153,
|
| 337 |
+
"grad_norm": 0.7455137586802161,
|
| 338 |
+
"learning_rate": 5e-06,
|
| 339 |
+
"loss": 0.5574,
|
| 340 |
+
"step": 450
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.078420310296192,
|
| 344 |
+
"grad_norm": 0.7317716271008468,
|
| 345 |
+
"learning_rate": 5e-06,
|
| 346 |
+
"loss": 0.5064,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.1235543018335683,
|
| 351 |
+
"grad_norm": 0.7140712178378773,
|
| 352 |
+
"learning_rate": 5e-06,
|
| 353 |
+
"loss": 0.5109,
|
| 354 |
+
"step": 470
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.168688293370945,
|
| 358 |
+
"grad_norm": 0.5219136342087098,
|
| 359 |
+
"learning_rate": 5e-06,
|
| 360 |
+
"loss": 0.5119,
|
| 361 |
+
"step": 480
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.213822284908322,
|
| 365 |
+
"grad_norm": 0.6942766993148072,
|
| 366 |
+
"learning_rate": 5e-06,
|
| 367 |
+
"loss": 0.5045,
|
| 368 |
+
"step": 490
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.2589562764456983,
|
| 372 |
+
"grad_norm": 0.4751025500905717,
|
| 373 |
+
"learning_rate": 5e-06,
|
| 374 |
+
"loss": 0.5058,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.304090267983075,
|
| 379 |
+
"grad_norm": 0.5526788063335546,
|
| 380 |
+
"learning_rate": 5e-06,
|
| 381 |
+
"loss": 0.5134,
|
| 382 |
+
"step": 510
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.3492242595204513,
|
| 386 |
+
"grad_norm": 0.5471049911581812,
|
| 387 |
+
"learning_rate": 5e-06,
|
| 388 |
+
"loss": 0.5135,
|
| 389 |
+
"step": 520
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.394358251057828,
|
| 393 |
+
"grad_norm": 0.5542894654716121,
|
| 394 |
+
"learning_rate": 5e-06,
|
| 395 |
+
"loss": 0.5115,
|
| 396 |
+
"step": 530
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.4394922425952044,
|
| 400 |
+
"grad_norm": 0.5478659351590376,
|
| 401 |
+
"learning_rate": 5e-06,
|
| 402 |
+
"loss": 0.5102,
|
| 403 |
+
"step": 540
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.4846262341325813,
|
| 407 |
+
"grad_norm": 0.5494840585563637,
|
| 408 |
+
"learning_rate": 5e-06,
|
| 409 |
+
"loss": 0.5163,
|
| 410 |
+
"step": 550
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.529760225669958,
|
| 414 |
+
"grad_norm": 0.49237680882767904,
|
| 415 |
+
"learning_rate": 5e-06,
|
| 416 |
+
"loss": 0.5115,
|
| 417 |
+
"step": 560
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.5748942172073344,
|
| 421 |
+
"grad_norm": 0.47486464629996683,
|
| 422 |
+
"learning_rate": 5e-06,
|
| 423 |
+
"loss": 0.5129,
|
| 424 |
+
"step": 570
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.620028208744711,
|
| 428 |
+
"grad_norm": 0.5165043118582577,
|
| 429 |
+
"learning_rate": 5e-06,
|
| 430 |
+
"loss": 0.5121,
|
| 431 |
+
"step": 580
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.6651622002820874,
|
| 435 |
+
"grad_norm": 0.5794476922681516,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.5117,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.710296191819464,
|
| 442 |
+
"grad_norm": 0.45233116281434377,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.5162,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.7554301833568404,
|
| 449 |
+
"grad_norm": 0.5214525369766981,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.5132,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.800564174894217,
|
| 456 |
+
"grad_norm": 0.5938791442315001,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.518,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.845698166431594,
|
| 463 |
+
"grad_norm": 0.5796881742506971,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.5195,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.8908321579689704,
|
| 470 |
+
"grad_norm": 0.47397608879546316,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.5135,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.935966149506347,
|
| 477 |
+
"grad_norm": 0.471420002781148,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.5154,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.9811001410437235,
|
| 484 |
+
"grad_norm": 0.45834788024298295,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.5153,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.9946403385049365,
|
| 491 |
+
"eval_loss": 0.5917297005653381,
|
| 492 |
+
"eval_runtime": 356.1014,
|
| 493 |
+
"eval_samples_per_second": 33.533,
|
| 494 |
+
"eval_steps_per_second": 0.525,
|
| 495 |
+
"step": 663
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.9946403385049365,
|
| 499 |
+
"step": 663,
|
| 500 |
+
"total_flos": 5052524767739904.0,
|
| 501 |
+
"train_loss": 0.5641303292586612,
|
| 502 |
+
"train_runtime": 61418.5213,
|
| 503 |
+
"train_samples_per_second": 11.081,
|
| 504 |
+
"train_steps_per_second": 0.011
|
| 505 |
+
}
|
| 506 |
+
],
|
| 507 |
+
"logging_steps": 10,
|
| 508 |
+
"max_steps": 663,
|
| 509 |
+
"num_input_tokens_seen": 0,
|
| 510 |
+
"num_train_epochs": 3,
|
| 511 |
+
"save_steps": 500,
|
| 512 |
+
"stateful_callbacks": {
|
| 513 |
+
"TrainerControl": {
|
| 514 |
+
"args": {
|
| 515 |
+
"should_epoch_stop": false,
|
| 516 |
+
"should_evaluate": false,
|
| 517 |
+
"should_log": false,
|
| 518 |
+
"should_save": true,
|
| 519 |
+
"should_training_stop": true
|
| 520 |
+
},
|
| 521 |
+
"attributes": {}
|
| 522 |
+
}
|
| 523 |
+
},
|
| 524 |
+
"total_flos": 5052524767739904.0,
|
| 525 |
+
"train_batch_size": 4,
|
| 526 |
+
"trial_name": null,
|
| 527 |
+
"trial_params": null
|
| 528 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|