checkpoint-360/trainer_state.json

{
 "best_metric": null,
 "best_model_checkpoint": null,
 "epoch": 2.981366459627329,
 "eval_steps": 500,
 "global_step": 360,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.2070393374741201,
 "grad_norm": 0.6150813102722168,
 "learning_rate": 0.0001388888888888889,
 "loss": 1.2294,
 "step": 25
 },
 {
 "epoch": 0.4140786749482402,
 "grad_norm": 0.6390131115913391,
 "learning_rate": 0.00019135802469135804,
 "loss": 0.7522,
 "step": 50
 },
 {
 "epoch": 0.6211180124223602,
 "grad_norm": 1.1554582118988037,
 "learning_rate": 0.00017592592592592595,
 "loss": 0.5117,
 "step": 75
 },
 {
 "epoch": 0.8281573498964804,
 "grad_norm": 0.4650109112262726,
 "learning_rate": 0.00016049382716049385,
 "loss": 0.4801,
 "step": 100
 },
 {
 "epoch": 1.0351966873706004,
 "grad_norm": 0.4381057024002075,
 "learning_rate": 0.00014506172839506173,
 "loss": 0.4299,
 "step": 125
 },
 {
 "epoch": 1.2422360248447206,
 "grad_norm": 0.5832701921463013,
 "learning_rate": 0.00012962962962962963,
 "loss": 0.4302,
 "step": 150
 },
 {
 "epoch": 1.4492753623188406,
 "grad_norm": 0.5579822659492493,
 "learning_rate": 0.00011419753086419754,
 "loss": 0.3808,
 "step": 175
 },
 {
 "epoch": 1.6563146997929605,
 "grad_norm": 0.4348982274532318,
 "learning_rate": 9.876543209876543e-05,
 "loss": 0.3839,
 "step": 200
 },
 {
 "epoch": 1.8633540372670807,
 "grad_norm": 0.661745548248291,
 "learning_rate": 8.333333333333334e-05,
 "loss": 0.4496,
 "step": 225
 },
 {
 "epoch": 2.070393374741201,
 "grad_norm": 0.47886422276496887,
 "learning_rate": 6.790123456790123e-05,
 "loss": 0.4054,
 "step": 250
 },
 {
 "epoch": 2.277432712215321,
 "grad_norm": 0.5070476531982422,
 "learning_rate": 5.246913580246914e-05,
 "loss": 0.4053,
 "step": 275
 },
 {
 "epoch": 2.4844720496894412,
 "grad_norm": 0.40826812386512756,
 "learning_rate": 3.7037037037037037e-05,
 "loss": 0.3673,
 "step": 300
 },
 {
 "epoch": 2.691511387163561,
 "grad_norm": 0.6191437840461731,
 "learning_rate": 2.1604938271604937e-05,
 "loss": 0.3548,
 "step": 325
 },
 {
 "epoch": 2.898550724637681,
 "grad_norm": 0.7395160794258118,
 "learning_rate": 6.172839506172839e-06,
 "loss": 0.3762,
 "step": 350
 }
 ],
 "logging_steps": 25,
 "max_steps": 360,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 3,
 "save_steps": 500,
 "total_flos": 6.645907168690176e+16,
 "train_batch_size": 1,
 "trial_name": null,
 "trial_params": null
}