| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9830508474576272, | |
| "eval_steps": 8, | |
| "global_step": 29, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03389830508474576, | |
| "grad_norm": 0.38113964825930874, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.1589, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03389830508474576, | |
| "eval_loss": 1.130394697189331, | |
| "eval_runtime": 17.254, | |
| "eval_samples_per_second": 5.738, | |
| "eval_steps_per_second": 0.406, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06779661016949153, | |
| "grad_norm": 0.3402704961945935, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.2375, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1016949152542373, | |
| "grad_norm": 0.3059822893552262, | |
| "learning_rate": 3e-06, | |
| "loss": 1.1753, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13559322033898305, | |
| "grad_norm": 0.2967260032476386, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.2375, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1694915254237288, | |
| "grad_norm": 0.34744521968923137, | |
| "learning_rate": 5e-06, | |
| "loss": 1.2547, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2033898305084746, | |
| "grad_norm": 0.35347423132256417, | |
| "learning_rate": 6e-06, | |
| "loss": 1.1697, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23728813559322035, | |
| "grad_norm": 0.3384189211692584, | |
| "learning_rate": 7e-06, | |
| "loss": 1.1887, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2711864406779661, | |
| "grad_norm": 0.30721189309671465, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.1859, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2711864406779661, | |
| "eval_loss": 1.128434181213379, | |
| "eval_runtime": 15.9228, | |
| "eval_samples_per_second": 6.218, | |
| "eval_steps_per_second": 0.44, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3050847457627119, | |
| "grad_norm": 0.4453319451582458, | |
| "learning_rate": 9e-06, | |
| "loss": 1.1759, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3389830508474576, | |
| "grad_norm": 0.3351797755126229, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2088, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3728813559322034, | |
| "grad_norm": 0.3941569965942398, | |
| "learning_rate": 9.931806517013612e-06, | |
| "loss": 1.207, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4067796610169492, | |
| "grad_norm": 0.3441878205899112, | |
| "learning_rate": 9.729086208503174e-06, | |
| "loss": 1.1378, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4406779661016949, | |
| "grad_norm": 0.41959508213449587, | |
| "learning_rate": 9.397368756032445e-06, | |
| "loss": 1.1803, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.4745762711864407, | |
| "grad_norm": 0.3756216639700884, | |
| "learning_rate": 8.94570254698197e-06, | |
| "loss": 1.1101, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5084745762711864, | |
| "grad_norm": 0.43480517409238956, | |
| "learning_rate": 8.386407858128707e-06, | |
| "loss": 1.217, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5423728813559322, | |
| "grad_norm": 0.4491620141825034, | |
| "learning_rate": 7.734740790612137e-06, | |
| "loss": 1.0445, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5423728813559322, | |
| "eval_loss": 1.1105434894561768, | |
| "eval_runtime": 15.8633, | |
| "eval_samples_per_second": 6.241, | |
| "eval_steps_per_second": 0.441, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.576271186440678, | |
| "grad_norm": 0.4666669843609912, | |
| "learning_rate": 7.008477123264849e-06, | |
| "loss": 1.2094, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6101694915254238, | |
| "grad_norm": 0.3831948927984007, | |
| "learning_rate": 6.227427435703997e-06, | |
| "loss": 1.1375, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6440677966101694, | |
| "grad_norm": 0.4141983253111484, | |
| "learning_rate": 5.412896727361663e-06, | |
| "loss": 1.1568, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 0.4528810988014663, | |
| "learning_rate": 4.587103272638339e-06, | |
| "loss": 1.0813, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.711864406779661, | |
| "grad_norm": 0.463559725998681, | |
| "learning_rate": 3.7725725642960047e-06, | |
| "loss": 1.0759, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7457627118644068, | |
| "grad_norm": 0.5481602130859397, | |
| "learning_rate": 2.991522876735154e-06, | |
| "loss": 1.1317, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.7796610169491526, | |
| "grad_norm": 0.4227477837808019, | |
| "learning_rate": 2.265259209387867e-06, | |
| "loss": 1.1669, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8135593220338984, | |
| "grad_norm": 0.42412143974912303, | |
| "learning_rate": 1.6135921418712959e-06, | |
| "loss": 1.1946, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8135593220338984, | |
| "eval_loss": 1.0844968557357788, | |
| "eval_runtime": 15.8637, | |
| "eval_samples_per_second": 6.241, | |
| "eval_steps_per_second": 0.441, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.847457627118644, | |
| "grad_norm": 0.42758121975218416, | |
| "learning_rate": 1.0542974530180327e-06, | |
| "loss": 1.174, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8813559322033898, | |
| "grad_norm": 0.5245615771378881, | |
| "learning_rate": 6.026312439675553e-07, | |
| "loss": 1.1217, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.9152542372881356, | |
| "grad_norm": 0.36119220018810566, | |
| "learning_rate": 2.7091379149682683e-07, | |
| "loss": 1.0711, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9491525423728814, | |
| "grad_norm": 0.42475966514744157, | |
| "learning_rate": 6.819348298638839e-08, | |
| "loss": 0.9551, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9830508474576272, | |
| "grad_norm": 0.4061658497077543, | |
| "learning_rate": 0.0, | |
| "loss": 1.1796, | |
| "step": 29 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 29, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 281581411368960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |