Commit
•
48d80b5
1
Parent(s):
3e3886a
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 +612 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: llama_8b_lima_8
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama_8b_lima_8
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.7044
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: llama_8b_lima_8
|
|
|
|
| 16 |
|
| 17 |
# llama_8b_lima_8
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the open_webui_dataset dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.7044
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.7043666839599609,
|
| 4 |
+
"eval_runtime": 8.0832,
|
| 5 |
+
"eval_samples_per_second": 27.217,
|
| 6 |
+
"eval_steps_per_second": 4.577,
|
| 7 |
+
"total_flos": 2.093861581933773e+16,
|
| 8 |
+
"train_loss": 0.6267224325452532,
|
| 9 |
+
"train_runtime": 1749.357,
|
| 10 |
+
"train_samples_per_second": 3.601,
|
| 11 |
+
"train_steps_per_second": 0.2
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.7043666839599609,
|
| 4 |
+
"eval_runtime": 8.0832,
|
| 5 |
+
"eval_samples_per_second": 27.217,
|
| 6 |
+
"eval_steps_per_second": 4.577
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.093861581933773e+16,
|
| 4 |
+
"train_loss": 0.6267224325452532,
|
| 5 |
+
"train_runtime": 1749.357,
|
| 6 |
+
"train_samples_per_second": 3.601,
|
| 7 |
+
"train_steps_per_second": 0.2
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,612 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 35,
|
| 6 |
+
"global_step": 350,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.014285714285714285,
|
| 13 |
+
"grad_norm": 118.65539307904277,
|
| 14 |
+
"learning_rate": 1.6e-06,
|
| 15 |
+
"loss": 0.8722,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02857142857142857,
|
| 20 |
+
"grad_norm": 38.484858029114484,
|
| 21 |
+
"learning_rate": 3.2e-06,
|
| 22 |
+
"loss": 0.7825,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04285714285714286,
|
| 27 |
+
"grad_norm": 88.4372189754248,
|
| 28 |
+
"learning_rate": 4.8e-06,
|
| 29 |
+
"loss": 0.7646,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05714285714285714,
|
| 34 |
+
"grad_norm": 41.455396892104744,
|
| 35 |
+
"learning_rate": 6.4e-06,
|
| 36 |
+
"loss": 0.7226,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07142857142857142,
|
| 41 |
+
"grad_norm": 4.738758549704674,
|
| 42 |
+
"learning_rate": 8e-06,
|
| 43 |
+
"loss": 0.6666,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08571428571428572,
|
| 48 |
+
"grad_norm": 10.5176134179179,
|
| 49 |
+
"learning_rate": 7.771005917159763e-06,
|
| 50 |
+
"loss": 0.7263,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1,
|
| 55 |
+
"grad_norm": 11.488275336430554,
|
| 56 |
+
"learning_rate": 7.545562130177514e-06,
|
| 57 |
+
"loss": 0.6211,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1,
|
| 62 |
+
"eval_loss": 0.9106850624084473,
|
| 63 |
+
"eval_runtime": 8.373,
|
| 64 |
+
"eval_samples_per_second": 26.275,
|
| 65 |
+
"eval_steps_per_second": 4.419,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11428571428571428,
|
| 70 |
+
"grad_norm": 18.92714653332412,
|
| 71 |
+
"learning_rate": 7.323668639053254e-06,
|
| 72 |
+
"loss": 0.6989,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.12857142857142856,
|
| 77 |
+
"grad_norm": 22.493765084792706,
|
| 78 |
+
"learning_rate": 7.105325443786982e-06,
|
| 79 |
+
"loss": 0.704,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.14285714285714285,
|
| 84 |
+
"grad_norm": 211.59677798517507,
|
| 85 |
+
"learning_rate": 6.890532544378699e-06,
|
| 86 |
+
"loss": 0.6654,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.15714285714285714,
|
| 91 |
+
"grad_norm": 160.57396424563427,
|
| 92 |
+
"learning_rate": 6.679289940828402e-06,
|
| 93 |
+
"loss": 0.7127,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.17142857142857143,
|
| 98 |
+
"grad_norm": 8.199790596273626,
|
| 99 |
+
"learning_rate": 6.471597633136094e-06,
|
| 100 |
+
"loss": 0.6253,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.18571428571428572,
|
| 105 |
+
"grad_norm": 50.8343087092876,
|
| 106 |
+
"learning_rate": 6.2674556213017745e-06,
|
| 107 |
+
"loss": 0.6973,
|
| 108 |
+
"step": 65
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.2,
|
| 112 |
+
"grad_norm": 99.89969831226269,
|
| 113 |
+
"learning_rate": 6.066863905325444e-06,
|
| 114 |
+
"loss": 0.6976,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.2,
|
| 119 |
+
"eval_loss": 0.852588951587677,
|
| 120 |
+
"eval_runtime": 8.1201,
|
| 121 |
+
"eval_samples_per_second": 27.093,
|
| 122 |
+
"eval_steps_per_second": 4.557,
|
| 123 |
+
"step": 70
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.21428571428571427,
|
| 127 |
+
"grad_norm": 68.18379393534644,
|
| 128 |
+
"learning_rate": 5.869822485207101e-06,
|
| 129 |
+
"loss": 0.6194,
|
| 130 |
+
"step": 75
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.22857142857142856,
|
| 134 |
+
"grad_norm": 10.735154673670019,
|
| 135 |
+
"learning_rate": 5.676331360946745e-06,
|
| 136 |
+
"loss": 0.6906,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.24285714285714285,
|
| 141 |
+
"grad_norm": 9.131803561236161,
|
| 142 |
+
"learning_rate": 5.486390532544378e-06,
|
| 143 |
+
"loss": 0.6401,
|
| 144 |
+
"step": 85
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.2571428571428571,
|
| 148 |
+
"grad_norm": 3.7707612659287766,
|
| 149 |
+
"learning_rate": 5.300000000000002e-06,
|
| 150 |
+
"loss": 0.6326,
|
| 151 |
+
"step": 90
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.2714285714285714,
|
| 155 |
+
"grad_norm": 8.70600675802851,
|
| 156 |
+
"learning_rate": 5.1171597633136094e-06,
|
| 157 |
+
"loss": 0.6019,
|
| 158 |
+
"step": 95
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.2857142857142857,
|
| 162 |
+
"grad_norm": 7.499128841456364,
|
| 163 |
+
"learning_rate": 4.9378698224852065e-06,
|
| 164 |
+
"loss": 0.6404,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.3,
|
| 169 |
+
"grad_norm": 5.203399746039487,
|
| 170 |
+
"learning_rate": 4.762130177514793e-06,
|
| 171 |
+
"loss": 0.5762,
|
| 172 |
+
"step": 105
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.3,
|
| 176 |
+
"eval_loss": 0.7999083399772644,
|
| 177 |
+
"eval_runtime": 8.0634,
|
| 178 |
+
"eval_samples_per_second": 27.284,
|
| 179 |
+
"eval_steps_per_second": 4.589,
|
| 180 |
+
"step": 105
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.3142857142857143,
|
| 184 |
+
"grad_norm": 4.900592118020277,
|
| 185 |
+
"learning_rate": 4.589940828402367e-06,
|
| 186 |
+
"loss": 0.7544,
|
| 187 |
+
"step": 110
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.32857142857142857,
|
| 191 |
+
"grad_norm": 6.583803116646508,
|
| 192 |
+
"learning_rate": 4.421301775147928e-06,
|
| 193 |
+
"loss": 0.6494,
|
| 194 |
+
"step": 115
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.34285714285714286,
|
| 198 |
+
"grad_norm": 2.8434894878977968,
|
| 199 |
+
"learning_rate": 4.2562130177514784e-06,
|
| 200 |
+
"loss": 0.6542,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.35714285714285715,
|
| 205 |
+
"grad_norm": 3.3996307039750926,
|
| 206 |
+
"learning_rate": 4.094674556213017e-06,
|
| 207 |
+
"loss": 0.6631,
|
| 208 |
+
"step": 125
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.37142857142857144,
|
| 212 |
+
"grad_norm": 5.094034483082871,
|
| 213 |
+
"learning_rate": 3.936686390532545e-06,
|
| 214 |
+
"loss": 0.6086,
|
| 215 |
+
"step": 130
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.38571428571428573,
|
| 219 |
+
"grad_norm": 3.8103990086460757,
|
| 220 |
+
"learning_rate": 3.7822485207100586e-06,
|
| 221 |
+
"loss": 0.5935,
|
| 222 |
+
"step": 135
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.4,
|
| 226 |
+
"grad_norm": 4.098923421731261,
|
| 227 |
+
"learning_rate": 3.631360946745561e-06,
|
| 228 |
+
"loss": 0.5226,
|
| 229 |
+
"step": 140
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.4,
|
| 233 |
+
"eval_loss": 0.7717307209968567,
|
| 234 |
+
"eval_runtime": 8.04,
|
| 235 |
+
"eval_samples_per_second": 27.363,
|
| 236 |
+
"eval_steps_per_second": 4.602,
|
| 237 |
+
"step": 140
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.4142857142857143,
|
| 241 |
+
"grad_norm": 3.3521626999923875,
|
| 242 |
+
"learning_rate": 3.484023668639053e-06,
|
| 243 |
+
"loss": 0.5967,
|
| 244 |
+
"step": 145
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.42857142857142855,
|
| 248 |
+
"grad_norm": 4.761409736989179,
|
| 249 |
+
"learning_rate": 3.3402366863905327e-06,
|
| 250 |
+
"loss": 0.6379,
|
| 251 |
+
"step": 150
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.44285714285714284,
|
| 255 |
+
"grad_norm": 5.189823443764368,
|
| 256 |
+
"learning_rate": 3.1999999999999994e-06,
|
| 257 |
+
"loss": 0.69,
|
| 258 |
+
"step": 155
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.45714285714285713,
|
| 262 |
+
"grad_norm": 3.4259629095344963,
|
| 263 |
+
"learning_rate": 3.0633136094674547e-06,
|
| 264 |
+
"loss": 0.6177,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.4714285714285714,
|
| 269 |
+
"grad_norm": 6.564952365844063,
|
| 270 |
+
"learning_rate": 2.930177514792899e-06,
|
| 271 |
+
"loss": 0.6451,
|
| 272 |
+
"step": 165
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.4857142857142857,
|
| 276 |
+
"grad_norm": 3.8605502771298346,
|
| 277 |
+
"learning_rate": 2.8005917159763313e-06,
|
| 278 |
+
"loss": 0.5706,
|
| 279 |
+
"step": 170
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5,
|
| 283 |
+
"grad_norm": 4.052127550985865,
|
| 284 |
+
"learning_rate": 2.674556213017751e-06,
|
| 285 |
+
"loss": 0.5866,
|
| 286 |
+
"step": 175
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.5,
|
| 290 |
+
"eval_loss": 0.7432886958122253,
|
| 291 |
+
"eval_runtime": 8.0548,
|
| 292 |
+
"eval_samples_per_second": 27.313,
|
| 293 |
+
"eval_steps_per_second": 4.594,
|
| 294 |
+
"step": 175
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5142857142857142,
|
| 298 |
+
"grad_norm": 3.3881437035921675,
|
| 299 |
+
"learning_rate": 2.5520710059171586e-06,
|
| 300 |
+
"loss": 0.6236,
|
| 301 |
+
"step": 180
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.5285714285714286,
|
| 305 |
+
"grad_norm": 3.6324500058620783,
|
| 306 |
+
"learning_rate": 2.4331360946745558e-06,
|
| 307 |
+
"loss": 0.614,
|
| 308 |
+
"step": 185
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.5428571428571428,
|
| 312 |
+
"grad_norm": 3.6721733069689377,
|
| 313 |
+
"learning_rate": 2.317751479289941e-06,
|
| 314 |
+
"loss": 0.5696,
|
| 315 |
+
"step": 190
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.5571428571428572,
|
| 319 |
+
"grad_norm": 3.2065483170410944,
|
| 320 |
+
"learning_rate": 2.2059171597633135e-06,
|
| 321 |
+
"loss": 0.6017,
|
| 322 |
+
"step": 195
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.5714285714285714,
|
| 326 |
+
"grad_norm": 3.694107283298511,
|
| 327 |
+
"learning_rate": 2.0976331360946745e-06,
|
| 328 |
+
"loss": 0.5215,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.5857142857142857,
|
| 333 |
+
"grad_norm": 3.3237350491778606,
|
| 334 |
+
"learning_rate": 1.9928994082840233e-06,
|
| 335 |
+
"loss": 0.6335,
|
| 336 |
+
"step": 205
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.6,
|
| 340 |
+
"grad_norm": 5.7877669531548595,
|
| 341 |
+
"learning_rate": 1.891715976331361e-06,
|
| 342 |
+
"loss": 0.5999,
|
| 343 |
+
"step": 210
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.6,
|
| 347 |
+
"eval_loss": 0.7398412227630615,
|
| 348 |
+
"eval_runtime": 8.0412,
|
| 349 |
+
"eval_samples_per_second": 27.359,
|
| 350 |
+
"eval_steps_per_second": 4.601,
|
| 351 |
+
"step": 210
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.6142857142857143,
|
| 355 |
+
"grad_norm": 4.758098459635445,
|
| 356 |
+
"learning_rate": 1.7940828402366863e-06,
|
| 357 |
+
"loss": 0.6837,
|
| 358 |
+
"step": 215
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.6285714285714286,
|
| 362 |
+
"grad_norm": 2.672565209825915,
|
| 363 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 364 |
+
"loss": 0.6247,
|
| 365 |
+
"step": 220
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.6428571428571429,
|
| 369 |
+
"grad_norm": 3.9151505842097594,
|
| 370 |
+
"learning_rate": 1.6094674556213014e-06,
|
| 371 |
+
"loss": 0.5437,
|
| 372 |
+
"step": 225
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.6571428571428571,
|
| 376 |
+
"grad_norm": 4.93388663600263,
|
| 377 |
+
"learning_rate": 1.5224852071005916e-06,
|
| 378 |
+
"loss": 0.6034,
|
| 379 |
+
"step": 230
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.6714285714285714,
|
| 383 |
+
"grad_norm": 2.9614002133596316,
|
| 384 |
+
"learning_rate": 1.4390532544378696e-06,
|
| 385 |
+
"loss": 0.5687,
|
| 386 |
+
"step": 235
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.6857142857142857,
|
| 390 |
+
"grad_norm": 3.738349914157848,
|
| 391 |
+
"learning_rate": 1.3591715976331362e-06,
|
| 392 |
+
"loss": 0.6636,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.7,
|
| 397 |
+
"grad_norm": 3.8772829249948852,
|
| 398 |
+
"learning_rate": 1.2828402366863903e-06,
|
| 399 |
+
"loss": 0.6464,
|
| 400 |
+
"step": 245
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.7,
|
| 404 |
+
"eval_loss": 0.7203609943389893,
|
| 405 |
+
"eval_runtime": 8.0382,
|
| 406 |
+
"eval_samples_per_second": 27.369,
|
| 407 |
+
"eval_steps_per_second": 4.603,
|
| 408 |
+
"step": 245
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7142857142857143,
|
| 412 |
+
"grad_norm": 2.972054074443536,
|
| 413 |
+
"learning_rate": 1.2100591715976333e-06,
|
| 414 |
+
"loss": 0.6081,
|
| 415 |
+
"step": 250
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7285714285714285,
|
| 419 |
+
"grad_norm": 2.8758679138381202,
|
| 420 |
+
"learning_rate": 1.1408284023668636e-06,
|
| 421 |
+
"loss": 0.5808,
|
| 422 |
+
"step": 255
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7428571428571429,
|
| 426 |
+
"grad_norm": 3.959990245370596,
|
| 427 |
+
"learning_rate": 1.0751479289940828e-06,
|
| 428 |
+
"loss": 0.6201,
|
| 429 |
+
"step": 260
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7571428571428571,
|
| 433 |
+
"grad_norm": 3.5665367608826632,
|
| 434 |
+
"learning_rate": 1.0130177514792898e-06,
|
| 435 |
+
"loss": 0.5034,
|
| 436 |
+
"step": 265
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7714285714285715,
|
| 440 |
+
"grad_norm": 4.028636856853669,
|
| 441 |
+
"learning_rate": 9.544378698224853e-07,
|
| 442 |
+
"loss": 0.5932,
|
| 443 |
+
"step": 270
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.7857142857142857,
|
| 447 |
+
"grad_norm": 4.258016842155138,
|
| 448 |
+
"learning_rate": 8.994082840236684e-07,
|
| 449 |
+
"loss": 0.5742,
|
| 450 |
+
"step": 275
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8,
|
| 454 |
+
"grad_norm": 3.5334205284517024,
|
| 455 |
+
"learning_rate": 8.479289940828401e-07,
|
| 456 |
+
"loss": 0.552,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8,
|
| 461 |
+
"eval_loss": 0.7106707692146301,
|
| 462 |
+
"eval_runtime": 8.068,
|
| 463 |
+
"eval_samples_per_second": 27.268,
|
| 464 |
+
"eval_steps_per_second": 4.586,
|
| 465 |
+
"step": 280
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8142857142857143,
|
| 469 |
+
"grad_norm": 6.70792314615095,
|
| 470 |
+
"learning_rate": 7.999999999999998e-07,
|
| 471 |
+
"loss": 0.6199,
|
| 472 |
+
"step": 285
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8285714285714286,
|
| 476 |
+
"grad_norm": 3.3423971423196983,
|
| 477 |
+
"learning_rate": 7.556213017751479e-07,
|
| 478 |
+
"loss": 0.5638,
|
| 479 |
+
"step": 290
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8428571428571429,
|
| 483 |
+
"grad_norm": 3.0750783422793497,
|
| 484 |
+
"learning_rate": 7.147928994082838e-07,
|
| 485 |
+
"loss": 0.5326,
|
| 486 |
+
"step": 295
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.8571428571428571,
|
| 490 |
+
"grad_norm": 4.7296945533927355,
|
| 491 |
+
"learning_rate": 6.775147928994083e-07,
|
| 492 |
+
"loss": 0.6127,
|
| 493 |
+
"step": 300
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8714285714285714,
|
| 497 |
+
"grad_norm": 3.833725884188028,
|
| 498 |
+
"learning_rate": 6.437869822485206e-07,
|
| 499 |
+
"loss": 0.5634,
|
| 500 |
+
"step": 305
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8857142857142857,
|
| 504 |
+
"grad_norm": 3.359034703004336,
|
| 505 |
+
"learning_rate": 6.136094674556213e-07,
|
| 506 |
+
"loss": 0.5592,
|
| 507 |
+
"step": 310
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.9,
|
| 511 |
+
"grad_norm": 3.1893425724586857,
|
| 512 |
+
"learning_rate": 5.869822485207099e-07,
|
| 513 |
+
"loss": 0.5657,
|
| 514 |
+
"step": 315
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9,
|
| 518 |
+
"eval_loss": 0.7130246758460999,
|
| 519 |
+
"eval_runtime": 8.0419,
|
| 520 |
+
"eval_samples_per_second": 27.357,
|
| 521 |
+
"eval_steps_per_second": 4.601,
|
| 522 |
+
"step": 315
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.9142857142857143,
|
| 526 |
+
"grad_norm": 3.3324510082581087,
|
| 527 |
+
"learning_rate": 5.63905325443787e-07,
|
| 528 |
+
"loss": 0.6271,
|
| 529 |
+
"step": 320
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.9285714285714286,
|
| 533 |
+
"grad_norm": 3.4362159022422736,
|
| 534 |
+
"learning_rate": 5.443786982248519e-07,
|
| 535 |
+
"loss": 0.5349,
|
| 536 |
+
"step": 325
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.9428571428571428,
|
| 540 |
+
"grad_norm": 4.804926939525648,
|
| 541 |
+
"learning_rate": 5.284023668639053e-07,
|
| 542 |
+
"loss": 0.7169,
|
| 543 |
+
"step": 330
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.9571428571428572,
|
| 547 |
+
"grad_norm": 4.408287391472689,
|
| 548 |
+
"learning_rate": 5.159763313609466e-07,
|
| 549 |
+
"loss": 0.606,
|
| 550 |
+
"step": 335
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.9714285714285714,
|
| 554 |
+
"grad_norm": 5.066493752587581,
|
| 555 |
+
"learning_rate": 5.071005917159763e-07,
|
| 556 |
+
"loss": 0.5669,
|
| 557 |
+
"step": 340
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.9857142857142858,
|
| 561 |
+
"grad_norm": 8.178486160006328,
|
| 562 |
+
"learning_rate": 5.01775147928994e-07,
|
| 563 |
+
"loss": 0.5541,
|
| 564 |
+
"step": 345
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.0,
|
| 568 |
+
"grad_norm": 3.7353925732066693,
|
| 569 |
+
"learning_rate": 5e-07,
|
| 570 |
+
"loss": 0.5687,
|
| 571 |
+
"step": 350
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.0,
|
| 575 |
+
"eval_loss": 0.7043666839599609,
|
| 576 |
+
"eval_runtime": 8.0404,
|
| 577 |
+
"eval_samples_per_second": 27.362,
|
| 578 |
+
"eval_steps_per_second": 4.602,
|
| 579 |
+
"step": 350
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.0,
|
| 583 |
+
"step": 350,
|
| 584 |
+
"total_flos": 2.093861581933773e+16,
|
| 585 |
+
"train_loss": 0.6267224325452532,
|
| 586 |
+
"train_runtime": 1749.357,
|
| 587 |
+
"train_samples_per_second": 3.601,
|
| 588 |
+
"train_steps_per_second": 0.2
|
| 589 |
+
}
|
| 590 |
+
],
|
| 591 |
+
"logging_steps": 5,
|
| 592 |
+
"max_steps": 350,
|
| 593 |
+
"num_input_tokens_seen": 0,
|
| 594 |
+
"num_train_epochs": 1,
|
| 595 |
+
"save_steps": 350,
|
| 596 |
+
"stateful_callbacks": {
|
| 597 |
+
"TrainerControl": {
|
| 598 |
+
"args": {
|
| 599 |
+
"should_epoch_stop": false,
|
| 600 |
+
"should_evaluate": false,
|
| 601 |
+
"should_log": false,
|
| 602 |
+
"should_save": true,
|
| 603 |
+
"should_training_stop": true
|
| 604 |
+
},
|
| 605 |
+
"attributes": {}
|
| 606 |
+
}
|
| 607 |
+
},
|
| 608 |
+
"total_flos": 2.093861581933773e+16,
|
| 609 |
+
"train_batch_size": 3,
|
| 610 |
+
"trial_name": null,
|
| 611 |
+
"trial_params": null
|
| 612 |
+
}
|