pythia-exp / trainer_state.json
TinyPixel's picture
Upload folder using huggingface_hub
650250c
raw
history blame
2.97 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9846153846153847,
"eval_steps": 500,
"global_step": 48,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 2.1776,
"step": 2
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 2.4224,
"step": 4
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 2.5657,
"step": 6
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 2.5934,
"step": 8
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 2.6175,
"step": 10
},
{
"epoch": 0.25,
"learning_rate": 0.0002,
"loss": 2.8582,
"step": 12
},
{
"epoch": 0.29,
"learning_rate": 0.0002,
"loss": 2.2232,
"step": 14
},
{
"epoch": 0.33,
"learning_rate": 0.0002,
"loss": 2.2464,
"step": 16
},
{
"epoch": 0.37,
"learning_rate": 0.0002,
"loss": 2.5335,
"step": 18
},
{
"epoch": 0.41,
"learning_rate": 0.0002,
"loss": 2.6144,
"step": 20
},
{
"epoch": 0.45,
"learning_rate": 0.0002,
"loss": 2.395,
"step": 22
},
{
"epoch": 0.49,
"learning_rate": 0.0002,
"loss": 2.8326,
"step": 24
},
{
"epoch": 0.53,
"learning_rate": 0.0002,
"loss": 2.3137,
"step": 26
},
{
"epoch": 0.57,
"learning_rate": 0.0002,
"loss": 2.4326,
"step": 28
},
{
"epoch": 0.62,
"learning_rate": 0.0002,
"loss": 2.5481,
"step": 30
},
{
"epoch": 0.66,
"learning_rate": 0.0002,
"loss": 2.3901,
"step": 32
},
{
"epoch": 0.7,
"learning_rate": 0.0002,
"loss": 2.4413,
"step": 34
},
{
"epoch": 0.74,
"learning_rate": 0.0002,
"loss": 2.6465,
"step": 36
},
{
"epoch": 0.78,
"learning_rate": 0.0002,
"loss": 2.2408,
"step": 38
},
{
"epoch": 0.82,
"learning_rate": 0.0002,
"loss": 2.1161,
"step": 40
},
{
"epoch": 0.86,
"learning_rate": 0.0002,
"loss": 2.4779,
"step": 42
},
{
"epoch": 0.9,
"learning_rate": 0.0002,
"loss": 2.3948,
"step": 44
},
{
"epoch": 0.94,
"learning_rate": 0.0002,
"loss": 2.6741,
"step": 46
},
{
"epoch": 0.98,
"learning_rate": 0.0002,
"loss": 2.584,
"step": 48
}
],
"logging_steps": 2,
"max_steps": 48,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 2045076543504384.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}