MHGanainy/best-performing-clustering-6
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +679 -0
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# best-performing-clustering-6
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 15 |
# best-performing-clustering-6
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 2.1335
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 2.1334645748138428,
|
| 4 |
+
"eval_runtime": 159.4929,
|
| 5 |
+
"eval_samples_per_second": 10.916,
|
| 6 |
+
"eval_steps_per_second": 1.367,
|
| 7 |
+
"perplexity": 8.444071308426771,
|
| 8 |
+
"total_flos": 1.6746639458304e+17,
|
| 9 |
+
"train_loss": 2.2412419762070965,
|
| 10 |
+
"train_runtime": 4787.9733,
|
| 11 |
+
"train_samples_per_second": 3.84,
|
| 12 |
+
"train_steps_per_second": 1.92
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 2.1334645748138428,
|
| 4 |
+
"eval_runtime": 159.4929,
|
| 5 |
+
"eval_samples_per_second": 10.916,
|
| 6 |
+
"eval_steps_per_second": 1.367,
|
| 7 |
+
"perplexity": 8.444071308426771
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.6746639458304e+17,
|
| 4 |
+
"train_loss": 2.2412419762070965,
|
| 5 |
+
"train_runtime": 4787.9733,
|
| 6 |
+
"train_samples_per_second": 3.84,
|
| 7 |
+
"train_steps_per_second": 1.92
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 9194,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010876658690450293,
|
| 13 |
+
"grad_norm": 0.11775965243577957,
|
| 14 |
+
"learning_rate": 6.666666666666667e-06,
|
| 15 |
+
"loss": 2.5177,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.021753317380900587,
|
| 20 |
+
"grad_norm": 0.21489302814006805,
|
| 21 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 22 |
+
"loss": 2.5141,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03262997607135088,
|
| 27 |
+
"grad_norm": 0.3868540823459625,
|
| 28 |
+
"learning_rate": 2e-05,
|
| 29 |
+
"loss": 2.4843,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04350663476180117,
|
| 34 |
+
"grad_norm": 0.40104538202285767,
|
| 35 |
+
"learning_rate": 1.999376221583229e-05,
|
| 36 |
+
"loss": 2.3961,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05438329345225147,
|
| 41 |
+
"grad_norm": 0.5164044499397278,
|
| 42 |
+
"learning_rate": 1.9975056645319428e-05,
|
| 43 |
+
"loss": 2.35,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06525995214270176,
|
| 48 |
+
"grad_norm": 0.47309133410453796,
|
| 49 |
+
"learning_rate": 1.9943906624723733e-05,
|
| 50 |
+
"loss": 2.3903,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07613661083315205,
|
| 55 |
+
"grad_norm": 0.5553125739097595,
|
| 56 |
+
"learning_rate": 1.990035101546626e-05,
|
| 57 |
+
"loss": 2.3195,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08701326952360235,
|
| 62 |
+
"grad_norm": 0.5993786454200745,
|
| 63 |
+
"learning_rate": 1.9844444155644974e-05,
|
| 64 |
+
"loss": 2.3576,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09788992821405264,
|
| 69 |
+
"grad_norm": 0.5128291845321655,
|
| 70 |
+
"learning_rate": 1.9776255792244897e-05,
|
| 71 |
+
"loss": 2.2987,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10876658690450294,
|
| 76 |
+
"grad_norm": 0.5672403573989868,
|
| 77 |
+
"learning_rate": 1.9695870994124752e-05,
|
| 78 |
+
"loss": 2.3408,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11964324559495323,
|
| 83 |
+
"grad_norm": 0.6869469285011292,
|
| 84 |
+
"learning_rate": 1.960339004588874e-05,
|
| 85 |
+
"loss": 2.3277,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.13051990428540353,
|
| 90 |
+
"grad_norm": 0.5485507249832153,
|
| 91 |
+
"learning_rate": 1.949892832277582e-05,
|
| 92 |
+
"loss": 2.3045,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.14139656297585382,
|
| 97 |
+
"grad_norm": 0.5480315089225769,
|
| 98 |
+
"learning_rate": 1.9382616146722495e-05,
|
| 99 |
+
"loss": 2.3185,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1522732216663041,
|
| 104 |
+
"grad_norm": 0.7362557649612427,
|
| 105 |
+
"learning_rate": 1.9254598623778825e-05,
|
| 106 |
+
"loss": 2.268,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1631498803567544,
|
| 111 |
+
"grad_norm": 0.6089996099472046,
|
| 112 |
+
"learning_rate": 1.911503546308037e-05,
|
| 113 |
+
"loss": 2.298,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.1740265390472047,
|
| 118 |
+
"grad_norm": 0.5844936966896057,
|
| 119 |
+
"learning_rate": 1.896410077760198e-05,
|
| 120 |
+
"loss": 2.2817,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.18490319773765498,
|
| 125 |
+
"grad_norm": 0.5984183549880981,
|
| 126 |
+
"learning_rate": 1.880198286694193e-05,
|
| 127 |
+
"loss": 2.2744,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.19577985642810528,
|
| 132 |
+
"grad_norm": 0.7692469358444214,
|
| 133 |
+
"learning_rate": 1.8628883982407505e-05,
|
| 134 |
+
"loss": 2.2415,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.20665651511855557,
|
| 139 |
+
"grad_norm": 0.6598834991455078,
|
| 140 |
+
"learning_rate": 1.8445020074694992e-05,
|
| 141 |
+
"loss": 2.2932,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2175331738090059,
|
| 146 |
+
"grad_norm": 0.8028424978256226,
|
| 147 |
+
"learning_rate": 1.8250620524478897e-05,
|
| 148 |
+
"loss": 2.2812,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.22840983249945618,
|
| 153 |
+
"grad_norm": 0.7637801766395569,
|
| 154 |
+
"learning_rate": 1.8045927856246524e-05,
|
| 155 |
+
"loss": 2.2787,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.23928649118990647,
|
| 160 |
+
"grad_norm": 0.6362522840499878,
|
| 161 |
+
"learning_rate": 1.7831197435734912e-05,
|
| 162 |
+
"loss": 2.2977,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.25016314988035676,
|
| 167 |
+
"grad_norm": 0.7283287048339844,
|
| 168 |
+
"learning_rate": 1.7606697151347526e-05,
|
| 169 |
+
"loss": 2.2645,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.26103980857080705,
|
| 174 |
+
"grad_norm": 0.9132130742073059,
|
| 175 |
+
"learning_rate": 1.7372707079948295e-05,
|
| 176 |
+
"loss": 2.2763,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.27191646726125734,
|
| 181 |
+
"grad_norm": 0.6510069966316223,
|
| 182 |
+
"learning_rate": 1.7129519137449776e-05,
|
| 183 |
+
"loss": 2.24,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.28279312595170764,
|
| 188 |
+
"grad_norm": 0.886200487613678,
|
| 189 |
+
"learning_rate": 1.6877436714631463e-05,
|
| 190 |
+
"loss": 2.2413,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2936697846421579,
|
| 195 |
+
"grad_norm": 0.784830629825592,
|
| 196 |
+
"learning_rate": 1.661677429864256e-05,
|
| 197 |
+
"loss": 2.2407,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3045464433326082,
|
| 202 |
+
"grad_norm": 0.7875649333000183,
|
| 203 |
+
"learning_rate": 1.6347857080661378e-05,
|
| 204 |
+
"loss": 2.2516,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3154231020230585,
|
| 209 |
+
"grad_norm": 0.7136234045028687,
|
| 210 |
+
"learning_rate": 1.607102055020087e-05,
|
| 211 |
+
"loss": 2.2033,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3262997607135088,
|
| 216 |
+
"grad_norm": 0.7511260509490967,
|
| 217 |
+
"learning_rate": 1.5786610076566388e-05,
|
| 218 |
+
"loss": 2.2971,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3371764194039591,
|
| 223 |
+
"grad_norm": 0.7906160354614258,
|
| 224 |
+
"learning_rate": 1.5494980477987843e-05,
|
| 225 |
+
"loss": 2.2292,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3480530780944094,
|
| 230 |
+
"grad_norm": 0.8098989129066467,
|
| 231 |
+
"learning_rate": 1.5196495578963808e-05,
|
| 232 |
+
"loss": 2.2046,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3589297367848597,
|
| 237 |
+
"grad_norm": 0.719405472278595,
|
| 238 |
+
"learning_rate": 1.4891527756369766e-05,
|
| 239 |
+
"loss": 2.2208,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.36980639547530997,
|
| 244 |
+
"grad_norm": 0.576866626739502,
|
| 245 |
+
"learning_rate": 1.4580457474896804e-05,
|
| 246 |
+
"loss": 2.2388,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.38068305416576026,
|
| 251 |
+
"grad_norm": 0.6755930185317993,
|
| 252 |
+
"learning_rate": 1.4263672812400284e-05,
|
| 253 |
+
"loss": 2.2581,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.39155971285621055,
|
| 258 |
+
"grad_norm": 0.7689741253852844,
|
| 259 |
+
"learning_rate": 1.394156897575067e-05,
|
| 260 |
+
"loss": 2.2315,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.40243637154666084,
|
| 265 |
+
"grad_norm": 0.7418312430381775,
|
| 266 |
+
"learning_rate": 1.3614547807790483e-05,
|
| 267 |
+
"loss": 2.3031,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.41331303023711113,
|
| 272 |
+
"grad_norm": 0.7450283765792847,
|
| 273 |
+
"learning_rate": 1.328301728601252e-05,
|
| 274 |
+
"loss": 2.2307,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4241896889275614,
|
| 279 |
+
"grad_norm": 0.7184864282608032,
|
| 280 |
+
"learning_rate": 1.2947391013584758e-05,
|
| 281 |
+
"loss": 2.2386,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.4350663476180118,
|
| 286 |
+
"grad_norm": 0.7286122441291809,
|
| 287 |
+
"learning_rate": 1.2608087703356877e-05,
|
| 288 |
+
"loss": 2.2103,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.44594300630846206,
|
| 293 |
+
"grad_norm": 0.9107334613800049,
|
| 294 |
+
"learning_rate": 1.22655306554922e-05,
|
| 295 |
+
"loss": 2.1938,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.45681966499891236,
|
| 300 |
+
"grad_norm": 0.8477912545204163,
|
| 301 |
+
"learning_rate": 1.1920147229376663e-05,
|
| 302 |
+
"loss": 2.1957,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.46769632368936265,
|
| 307 |
+
"grad_norm": 0.8039445281028748,
|
| 308 |
+
"learning_rate": 1.1572368310463712e-05,
|
| 309 |
+
"loss": 2.2234,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.47857298237981294,
|
| 314 |
+
"grad_norm": 0.7385637760162354,
|
| 315 |
+
"learning_rate": 1.1222627772720196e-05,
|
| 316 |
+
"loss": 2.2494,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.48944964107026323,
|
| 321 |
+
"grad_norm": 0.873935878276825,
|
| 322 |
+
"learning_rate": 1.0871361937343945e-05,
|
| 323 |
+
"loss": 2.209,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5003262997607135,
|
| 328 |
+
"grad_norm": 1.1287140846252441,
|
| 329 |
+
"learning_rate": 1.0519009028428272e-05,
|
| 330 |
+
"loss": 2.1784,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5112029584511638,
|
| 335 |
+
"grad_norm": 0.7948943376541138,
|
| 336 |
+
"learning_rate": 1.0166008626252514e-05,
|
| 337 |
+
"loss": 2.2169,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.5220796171416141,
|
| 342 |
+
"grad_norm": 0.75812828540802,
|
| 343 |
+
"learning_rate": 9.812801118880645e-06,
|
| 344 |
+
"loss": 2.2284,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5329562758320644,
|
| 349 |
+
"grad_norm": 1.0229548215866089,
|
| 350 |
+
"learning_rate": 9.459827152752154e-06,
|
| 351 |
+
"loss": 2.2066,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5438329345225147,
|
| 356 |
+
"grad_norm": 1.0001264810562134,
|
| 357 |
+
"learning_rate": 9.107527082950537e-06,
|
| 358 |
+
"loss": 2.2144,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.554709593212965,
|
| 363 |
+
"grad_norm": 1.3419371843338013,
|
| 364 |
+
"learning_rate": 8.756340423835338e-06,
|
| 365 |
+
"loss": 2.2282,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5655862519034153,
|
| 370 |
+
"grad_norm": 1.1223336458206177,
|
| 371 |
+
"learning_rate": 8.40670530072298e-06,
|
| 372 |
+
"loss": 2.1892,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5764629105938656,
|
| 377 |
+
"grad_norm": 0.6637461185455322,
|
| 378 |
+
"learning_rate": 8.059057903300553e-06,
|
| 379 |
+
"loss": 2.2495,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5873395692843159,
|
| 384 |
+
"grad_norm": 0.747543454170227,
|
| 385 |
+
"learning_rate": 7.713831941454373e-06,
|
| 386 |
+
"loss": 2.1836,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5982162279747661,
|
| 391 |
+
"grad_norm": 1.1951714754104614,
|
| 392 |
+
"learning_rate": 7.371458104192257e-06,
|
| 393 |
+
"loss": 2.1827,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.6090928866652164,
|
| 398 |
+
"grad_norm": 0.8208951354026794,
|
| 399 |
+
"learning_rate": 7.032363522334502e-06,
|
| 400 |
+
"loss": 2.2302,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.6199695453556667,
|
| 405 |
+
"grad_norm": 1.0612913370132446,
|
| 406 |
+
"learning_rate": 6.696971235643932e-06,
|
| 407 |
+
"loss": 2.164,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.630846204046117,
|
| 412 |
+
"grad_norm": 0.6853541135787964,
|
| 413 |
+
"learning_rate": 6.365699665059718e-06,
|
| 414 |
+
"loss": 2.1821,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6417228627365673,
|
| 419 |
+
"grad_norm": 0.8484651446342468,
|
| 420 |
+
"learning_rate": 6.038962090693503e-06,
|
| 421 |
+
"loss": 2.1819,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6525995214270176,
|
| 426 |
+
"grad_norm": 1.4542567729949951,
|
| 427 |
+
"learning_rate": 5.717166136238958e-06,
|
| 428 |
+
"loss": 2.1762,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6634761801174679,
|
| 433 |
+
"grad_norm": 1.2847181558609009,
|
| 434 |
+
"learning_rate": 5.400713260438077e-06,
|
| 435 |
+
"loss": 2.1901,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6743528388079182,
|
| 440 |
+
"grad_norm": 0.8455380201339722,
|
| 441 |
+
"learning_rate": 5.089998256238553e-06,
|
| 442 |
+
"loss": 2.1897,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6852294974983685,
|
| 447 |
+
"grad_norm": 1.1807805299758911,
|
| 448 |
+
"learning_rate": 4.785408758267164e-06,
|
| 449 |
+
"loss": 2.2193,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6961061561888188,
|
| 454 |
+
"grad_norm": 0.7739274501800537,
|
| 455 |
+
"learning_rate": 4.487324759233518e-06,
|
| 456 |
+
"loss": 2.2115,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.7069828148792691,
|
| 461 |
+
"grad_norm": 0.8319265246391296,
|
| 462 |
+
"learning_rate": 4.196118135867595e-06,
|
| 463 |
+
"loss": 2.178,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.7178594735697194,
|
| 468 |
+
"grad_norm": 0.8209885358810425,
|
| 469 |
+
"learning_rate": 3.912152184982336e-06,
|
| 470 |
+
"loss": 2.1665,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.7287361322601696,
|
| 475 |
+
"grad_norm": 1.5596791505813599,
|
| 476 |
+
"learning_rate": 3.6357811702402703e-06,
|
| 477 |
+
"loss": 2.232,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.7396127909506199,
|
| 482 |
+
"grad_norm": 0.8522664308547974,
|
| 483 |
+
"learning_rate": 3.3673498801894178e-06,
|
| 484 |
+
"loss": 2.1873,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7504894496410702,
|
| 489 |
+
"grad_norm": 0.7811763882637024,
|
| 490 |
+
"learning_rate": 3.1071931981200353e-06,
|
| 491 |
+
"loss": 2.1903,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7613661083315205,
|
| 496 |
+
"grad_norm": 1.0121145248413086,
|
| 497 |
+
"learning_rate": 2.855635684278618e-06,
|
| 498 |
+
"loss": 2.2129,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7722427670219708,
|
| 503 |
+
"grad_norm": 0.8791484832763672,
|
| 504 |
+
"learning_rate": 2.612991170960595e-06,
|
| 505 |
+
"loss": 2.1658,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7831194257124211,
|
| 510 |
+
"grad_norm": 1.054653525352478,
|
| 511 |
+
"learning_rate": 2.37956237098667e-06,
|
| 512 |
+
"loss": 2.1966,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7939960844028714,
|
| 517 |
+
"grad_norm": 0.8596212267875671,
|
| 518 |
+
"learning_rate": 2.1556405000514015e-06,
|
| 519 |
+
"loss": 2.2237,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.8048727430933217,
|
| 524 |
+
"grad_norm": 0.8690965175628662,
|
| 525 |
+
"learning_rate": 1.9415049134150545e-06,
|
| 526 |
+
"loss": 2.2035,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.815749401783772,
|
| 531 |
+
"grad_norm": 0.8708688020706177,
|
| 532 |
+
"learning_rate": 1.7374227573920377e-06,
|
| 533 |
+
"loss": 2.1783,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.8266260604742223,
|
| 538 |
+
"grad_norm": 0.8724594116210938,
|
| 539 |
+
"learning_rate": 1.5436486360707037e-06,
|
| 540 |
+
"loss": 2.2229,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.8375027191646726,
|
| 545 |
+
"grad_norm": 0.7637216448783875,
|
| 546 |
+
"learning_rate": 1.3604242936802702e-06,
|
| 547 |
+
"loss": 2.2131,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.8483793778551229,
|
| 552 |
+
"grad_norm": 0.8904792070388794,
|
| 553 |
+
"learning_rate": 1.1879783130011601e-06,
|
| 554 |
+
"loss": 2.2124,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8592560365455733,
|
| 559 |
+
"grad_norm": 0.920068085193634,
|
| 560 |
+
"learning_rate": 1.026525830194983e-06,
|
| 561 |
+
"loss": 2.2312,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.8701326952360235,
|
| 566 |
+
"grad_norm": 0.853566586971283,
|
| 567 |
+
"learning_rate": 8.762682664099564e-07,
|
| 568 |
+
"loss": 2.2061,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8810093539264738,
|
| 573 |
+
"grad_norm": 0.8593130111694336,
|
| 574 |
+
"learning_rate": 7.373930764965742e-07,
|
| 575 |
+
"loss": 2.1988,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8918860126169241,
|
| 580 |
+
"grad_norm": 0.8238704204559326,
|
| 581 |
+
"learning_rate": 6.100735151470216e-07,
|
| 582 |
+
"loss": 2.1747,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.9027626713073744,
|
| 587 |
+
"grad_norm": 1.1148990392684937,
|
| 588 |
+
"learning_rate": 4.944684207501005e-07,
|
| 589 |
+
"loss": 2.2112,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9136393299978247,
|
| 594 |
+
"grad_norm": 1.3240580558776855,
|
| 595 |
+
"learning_rate": 3.9072201723132084e-07,
|
| 596 |
+
"loss": 2.1655,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.924515988688275,
|
| 601 |
+
"grad_norm": 0.9755560755729675,
|
| 602 |
+
"learning_rate": 2.989637341253493e-07,
|
| 603 |
+
"loss": 2.1778,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.9353926473787253,
|
| 608 |
+
"grad_norm": 0.8737950921058655,
|
| 609 |
+
"learning_rate": 2.1930804510530536e-07,
|
| 610 |
+
"loss": 2.1974,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9462693060691756,
|
| 615 |
+
"grad_norm": 1.1282422542572021,
|
| 616 |
+
"learning_rate": 1.518543251703586e-07,
|
| 617 |
+
"loss": 2.2214,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9571459647596259,
|
| 622 |
+
"grad_norm": 0.8447544574737549,
|
| 623 |
+
"learning_rate": 9.66867266697602e-08,
|
| 624 |
+
"loss": 2.143,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9680226234500762,
|
| 629 |
+
"grad_norm": 0.9211781620979309,
|
| 630 |
+
"learning_rate": 5.387407431801217e-08,
|
| 631 |
+
"loss": 2.1945,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9788992821405265,
|
| 636 |
+
"grad_norm": 1.0062124729156494,
|
| 637 |
+
"learning_rate": 2.3469779332114806e-08,
|
| 638 |
+
"loss": 2.2168,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9897759408309768,
|
| 643 |
+
"grad_norm": 0.9540311694145203,
|
| 644 |
+
"learning_rate": 5.511772798049153e-09,
|
| 645 |
+
"loss": 2.186,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.0,
|
| 650 |
+
"step": 9194,
|
| 651 |
+
"total_flos": 1.6746639458304e+17,
|
| 652 |
+
"train_loss": 2.2412419762070965,
|
| 653 |
+
"train_runtime": 4787.9733,
|
| 654 |
+
"train_samples_per_second": 3.84,
|
| 655 |
+
"train_steps_per_second": 1.92
|
| 656 |
+
}
|
| 657 |
+
],
|
| 658 |
+
"logging_steps": 100,
|
| 659 |
+
"max_steps": 9194,
|
| 660 |
+
"num_input_tokens_seen": 0,
|
| 661 |
+
"num_train_epochs": 1,
|
| 662 |
+
"save_steps": 500,
|
| 663 |
+
"stateful_callbacks": {
|
| 664 |
+
"TrainerControl": {
|
| 665 |
+
"args": {
|
| 666 |
+
"should_epoch_stop": false,
|
| 667 |
+
"should_evaluate": false,
|
| 668 |
+
"should_log": false,
|
| 669 |
+
"should_save": true,
|
| 670 |
+
"should_training_stop": true
|
| 671 |
+
},
|
| 672 |
+
"attributes": {}
|
| 673 |
+
}
|
| 674 |
+
},
|
| 675 |
+
"total_flos": 1.6746639458304e+17,
|
| 676 |
+
"train_batch_size": 2,
|
| 677 |
+
"trial_name": null,
|
| 678 |
+
"trial_params": null
|
| 679 |
+
}
|