New model Version
Browse files- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +143 -0
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6991cc6552c5b76dbe1f0de75531c3a899fb22204ca73ad87b7212dd9296b6da
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a415030a49c53672ccac7723cf7f1e258eb6d2a6e1da2d25ddca5348e51f0041
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,143 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2674808204174042,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-2416",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7248,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.20695364238410596,
|
| 13 |
+
"grad_norm": 0.06496760994195938,
|
| 14 |
+
"learning_rate": 1.862030905077263e-05,
|
| 15 |
+
"loss": 0.1879,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.4139072847682119,
|
| 20 |
+
"grad_norm": 0.2046840637922287,
|
| 21 |
+
"learning_rate": 1.7240618101545256e-05,
|
| 22 |
+
"loss": 0.1544,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.6208609271523179,
|
| 27 |
+
"grad_norm": 78.30413818359375,
|
| 28 |
+
"learning_rate": 1.5860927152317882e-05,
|
| 29 |
+
"loss": 0.1443,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.8278145695364238,
|
| 34 |
+
"grad_norm": 13.444316864013672,
|
| 35 |
+
"learning_rate": 1.448123620309051e-05,
|
| 36 |
+
"loss": 0.1214,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 0.2674808204174042,
|
| 42 |
+
"eval_runtime": 68.0862,
|
| 43 |
+
"eval_samples_per_second": 31.548,
|
| 44 |
+
"eval_steps_per_second": 3.951,
|
| 45 |
+
"step": 2416
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.0347682119205297,
|
| 49 |
+
"grad_norm": 11.859747886657715,
|
| 50 |
+
"learning_rate": 1.3101545253863135e-05,
|
| 51 |
+
"loss": 0.1299,
|
| 52 |
+
"step": 2500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.2417218543046358,
|
| 56 |
+
"grad_norm": 7.801605701446533,
|
| 57 |
+
"learning_rate": 1.1721854304635763e-05,
|
| 58 |
+
"loss": 0.1358,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.4486754966887418,
|
| 63 |
+
"grad_norm": 9.417176246643066,
|
| 64 |
+
"learning_rate": 1.034216335540839e-05,
|
| 65 |
+
"loss": 0.1331,
|
| 66 |
+
"step": 3500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.6556291390728477,
|
| 70 |
+
"grad_norm": 0.017003627493977547,
|
| 71 |
+
"learning_rate": 8.962472406181017e-06,
|
| 72 |
+
"loss": 0.1325,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.8625827814569536,
|
| 77 |
+
"grad_norm": 7.609455585479736,
|
| 78 |
+
"learning_rate": 7.582781456953643e-06,
|
| 79 |
+
"loss": 0.1273,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.0,
|
| 84 |
+
"eval_loss": 0.2778768837451935,
|
| 85 |
+
"eval_runtime": 67.8572,
|
| 86 |
+
"eval_samples_per_second": 31.655,
|
| 87 |
+
"eval_steps_per_second": 3.964,
|
| 88 |
+
"step": 4832
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.0695364238410594,
|
| 92 |
+
"grad_norm": 0.0077050491236150265,
|
| 93 |
+
"learning_rate": 6.203090507726269e-06,
|
| 94 |
+
"loss": 0.125,
|
| 95 |
+
"step": 5000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.2764900662251657,
|
| 99 |
+
"grad_norm": 9.524530410766602,
|
| 100 |
+
"learning_rate": 4.823399558498897e-06,
|
| 101 |
+
"loss": 0.0672,
|
| 102 |
+
"step": 5500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.4834437086092715,
|
| 106 |
+
"grad_norm": 3.2392263412475586,
|
| 107 |
+
"learning_rate": 3.443708609271523e-06,
|
| 108 |
+
"loss": 0.0731,
|
| 109 |
+
"step": 6000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.6903973509933774,
|
| 113 |
+
"grad_norm": 0.4321236312389374,
|
| 114 |
+
"learning_rate": 2.06401766004415e-06,
|
| 115 |
+
"loss": 0.0818,
|
| 116 |
+
"step": 6500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.8973509933774837,
|
| 120 |
+
"grad_norm": 0.006676756776869297,
|
| 121 |
+
"learning_rate": 6.843267108167771e-07,
|
| 122 |
+
"loss": 0.0691,
|
| 123 |
+
"step": 7000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 3.0,
|
| 127 |
+
"eval_loss": 0.30005592107772827,
|
| 128 |
+
"eval_runtime": 67.9967,
|
| 129 |
+
"eval_samples_per_second": 31.59,
|
| 130 |
+
"eval_steps_per_second": 3.956,
|
| 131 |
+
"step": 7248
|
| 132 |
+
}
|
| 133 |
+
],
|
| 134 |
+
"logging_steps": 500,
|
| 135 |
+
"max_steps": 7248,
|
| 136 |
+
"num_input_tokens_seen": 0,
|
| 137 |
+
"num_train_epochs": 3,
|
| 138 |
+
"save_steps": 500,
|
| 139 |
+
"total_flos": 1.52529694887168e+16,
|
| 140 |
+
"train_batch_size": 8,
|
| 141 |
+
"trial_name": null,
|
| 142 |
+
"trial_params": null
|
| 143 |
+
}
|