Commit
•
dbddc7b
1
Parent(s):
b0102f0
Upload trainer_state.json
Browse files- trainer_state.json +140 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,140 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.88786658999425,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2875215641173088,
|
| 13 |
+
"grad_norm": 423.2573547363281,
|
| 14 |
+
"learning_rate": 4.712478435882691e-05,
|
| 15 |
+
"loss": 17.6035,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5750431282346176,
|
| 20 |
+
"grad_norm": 9.149832725524902,
|
| 21 |
+
"learning_rate": 4.4249568717653824e-05,
|
| 22 |
+
"loss": 2.5022,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.8625646923519263,
|
| 27 |
+
"grad_norm": 3.246171236038208,
|
| 28 |
+
"learning_rate": 4.137435307648074e-05,
|
| 29 |
+
"loss": 1.3833,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.1500862564692351,
|
| 34 |
+
"grad_norm": 5.592076778411865,
|
| 35 |
+
"learning_rate": 3.8499137435307647e-05,
|
| 36 |
+
"loss": 1.0705,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.437607820586544,
|
| 41 |
+
"grad_norm": 12.64777660369873,
|
| 42 |
+
"learning_rate": 3.562392179413456e-05,
|
| 43 |
+
"loss": 0.9577,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.7251293847038527,
|
| 48 |
+
"grad_norm": 0.7939251065254211,
|
| 49 |
+
"learning_rate": 3.2748706152961475e-05,
|
| 50 |
+
"loss": 0.919,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 2.0126509488211615,
|
| 55 |
+
"grad_norm": 0.4556220769882202,
|
| 56 |
+
"learning_rate": 2.9873490511788386e-05,
|
| 57 |
+
"loss": 0.8873,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 2.3001725129384702,
|
| 62 |
+
"grad_norm": 0.45749151706695557,
|
| 63 |
+
"learning_rate": 2.6998274870615297e-05,
|
| 64 |
+
"loss": 0.8553,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.587694077055779,
|
| 69 |
+
"grad_norm": 0.6427181363105774,
|
| 70 |
+
"learning_rate": 2.4123059229442212e-05,
|
| 71 |
+
"loss": 0.8427,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.875215641173088,
|
| 76 |
+
"grad_norm": 0.35242655873298645,
|
| 77 |
+
"learning_rate": 2.124784358826912e-05,
|
| 78 |
+
"loss": 0.8227,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 3.1627372052903966,
|
| 83 |
+
"grad_norm": 0.5690515041351318,
|
| 84 |
+
"learning_rate": 1.8372627947096034e-05,
|
| 85 |
+
"loss": 0.808,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.4502587694077054,
|
| 90 |
+
"grad_norm": 0.46968725323677063,
|
| 91 |
+
"learning_rate": 1.5497412305922945e-05,
|
| 92 |
+
"loss": 0.8198,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 3.7377803335250146,
|
| 97 |
+
"grad_norm": 0.4196397662162781,
|
| 98 |
+
"learning_rate": 1.2622196664749858e-05,
|
| 99 |
+
"loss": 0.8029,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.025301897642323,
|
| 104 |
+
"grad_norm": 0.5571260452270508,
|
| 105 |
+
"learning_rate": 9.746981023576769e-06,
|
| 106 |
+
"loss": 0.8049,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.312823461759632,
|
| 111 |
+
"grad_norm": 0.5123286247253418,
|
| 112 |
+
"learning_rate": 6.871765382403681e-06,
|
| 113 |
+
"loss": 0.7934,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 4.6003450258769405,
|
| 118 |
+
"grad_norm": 0.3241867423057556,
|
| 119 |
+
"learning_rate": 3.9965497412305925e-06,
|
| 120 |
+
"loss": 0.796,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 4.88786658999425,
|
| 125 |
+
"grad_norm": 0.4111732542514801,
|
| 126 |
+
"learning_rate": 1.1213341000575044e-06,
|
| 127 |
+
"loss": 0.7851,
|
| 128 |
+
"step": 8500
|
| 129 |
+
}
|
| 130 |
+
],
|
| 131 |
+
"logging_steps": 500,
|
| 132 |
+
"max_steps": 8695,
|
| 133 |
+
"num_input_tokens_seen": 0,
|
| 134 |
+
"num_train_epochs": 5,
|
| 135 |
+
"save_steps": 500,
|
| 136 |
+
"total_flos": 4492786123407360.0,
|
| 137 |
+
"train_batch_size": 8,
|
| 138 |
+
"trial_name": null,
|
| 139 |
+
"trial_params": null
|
| 140 |
+
}
|