| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.41322988975092, | |
| "eval_steps": 50, | |
| "global_step": 1900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 10.6083, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 10.407150268554688, | |
| "eval_runtime": 65.9544, | |
| "eval_samples_per_second": 0.561, | |
| "eval_steps_per_second": 0.561, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 10.0615, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 9.026963233947754, | |
| "eval_runtime": 66.0596, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 8.0748, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 6.740030288696289, | |
| "eval_runtime": 66.4335, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.999831241633323e-05, | |
| "loss": 6.2551, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 5.468978404998779, | |
| "eval_runtime": 66.4722, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.9939306773179498e-05, | |
| "loss": 5.3533, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 4.9109086990356445, | |
| "eval_runtime": 66.1035, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.979649067087574e-05, | |
| "loss": 4.9846, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 4.694249629974365, | |
| "eval_runtime": 66.2471, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.9571068366759143e-05, | |
| "loss": 4.8176, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": 4.578726768493652, | |
| "eval_runtime": 66.1659, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.9264940672148018e-05, | |
| "loss": 4.7153, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_loss": 4.5052080154418945, | |
| "eval_runtime": 66.4396, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.888068892427538e-05, | |
| "loss": 4.6504, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_loss": 4.450746536254883, | |
| "eval_runtime": 66.6083, | |
| "eval_samples_per_second": 0.555, | |
| "eval_steps_per_second": 0.555, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 1.842155321987566e-05, | |
| "loss": 4.5897, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "eval_loss": 4.411241054534912, | |
| "eval_runtime": 66.3951, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 1.789140509396394e-05, | |
| "loss": 4.5623, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_loss": 4.383650302886963, | |
| "eval_runtime": 66.4603, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.729471487418621e-05, | |
| "loss": 4.5371, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_loss": 4.359971523284912, | |
| "eval_runtime": 66.5261, | |
| "eval_samples_per_second": 0.556, | |
| "eval_steps_per_second": 0.556, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.6636513986016215e-05, | |
| "loss": 4.5189, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_loss": 4.340854167938232, | |
| "eval_runtime": 66.0697, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 1.5922352526649803e-05, | |
| "loss": 4.4797, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "eval_loss": 4.326164722442627, | |
| "eval_runtime": 66.0876, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 1.5158252465343242e-05, | |
| "loss": 4.4863, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_loss": 4.313417911529541, | |
| "eval_runtime": 66.3814, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 1.4350656864820733e-05, | |
| "loss": 4.4571, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "eval_loss": 4.301650047302246, | |
| "eval_runtime": 66.0901, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 1.3506375551927546e-05, | |
| "loss": 4.4562, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "eval_loss": 4.292283058166504, | |
| "eval_runtime": 66.2796, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.558, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1.2632527695645993e-05, | |
| "loss": 4.4527, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 4.286748886108398, | |
| "eval_runtime": 66.2321, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 4.4292, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 4.277628421783447, | |
| "eval_runtime": 66.6577, | |
| "eval_samples_per_second": 0.555, | |
| "eval_steps_per_second": 0.555, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 1.0825793454723325e-05, | |
| "loss": 4.423, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "eval_loss": 4.27248477935791, | |
| "eval_runtime": 66.2871, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.558, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 9.908141857552737e-06, | |
| "loss": 4.423, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "eval_loss": 4.266038417816162, | |
| "eval_runtime": 66.5179, | |
| "eval_samples_per_second": 0.556, | |
| "eval_steps_per_second": 0.556, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 8.991264828797319e-06, | |
| "loss": 4.4166, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "eval_loss": 4.2608160972595215, | |
| "eval_runtime": 66.1693, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 8.082893680762619e-06, | |
| "loss": 4.4156, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "eval_loss": 4.257768630981445, | |
| "eval_runtime": 66.5456, | |
| "eval_samples_per_second": 0.556, | |
| "eval_steps_per_second": 0.556, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 7.190688002264308e-06, | |
| "loss": 4.3988, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_loss": 4.253511428833008, | |
| "eval_runtime": 66.2492, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.558, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 6.322171071261071e-06, | |
| "loss": 4.4021, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "eval_loss": 4.251075744628906, | |
| "eval_runtime": 66.2478, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 5.484666416891109e-06, | |
| "loss": 4.4038, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "eval_loss": 4.247603893280029, | |
| "eval_runtime": 66.0732, | |
| "eval_samples_per_second": 0.56, | |
| "eval_steps_per_second": 0.56, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 4.685236065835443e-06, | |
| "loss": 4.3881, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "eval_loss": 4.24613094329834, | |
| "eval_runtime": 66.4204, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 3.930620993728434e-06, | |
| "loss": 4.3943, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "eval_loss": 4.2452192306518555, | |
| "eval_runtime": 66.4796, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 3.2271842837425917e-06, | |
| "loss": 4.3822, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "eval_loss": 4.24231481552124, | |
| "eval_runtime": 66.529, | |
| "eval_samples_per_second": 0.556, | |
| "eval_steps_per_second": 0.556, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 2.580857471647186e-06, | |
| "loss": 4.4064, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_loss": 4.241450309753418, | |
| "eval_runtime": 66.3417, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.558, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 1.9970905297711606e-06, | |
| "loss": 4.3793, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "eval_loss": 4.239691257476807, | |
| "eval_runtime": 66.2256, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "learning_rate": 1.4808059116167306e-06, | |
| "loss": 4.385, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "eval_loss": 4.239114761352539, | |
| "eval_runtime": 66.4764, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "learning_rate": 1.0363570446297999e-06, | |
| "loss": 4.3919, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "eval_loss": 4.239101409912109, | |
| "eval_runtime": 66.7261, | |
| "eval_samples_per_second": 0.555, | |
| "eval_steps_per_second": 0.555, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 6.67491621125429e-07, | |
| "loss": 4.3934, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 4.238786220550537, | |
| "eval_runtime": 66.2354, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 3.773199969074959e-07, | |
| "loss": 4.3825, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "eval_loss": 4.238570690155029, | |
| "eval_runtime": 66.3209, | |
| "eval_samples_per_second": 0.558, | |
| "eval_steps_per_second": 0.558, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 1.6828896405244988e-07, | |
| "loss": 4.3835, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 4.2383294105529785, | |
| "eval_runtime": 66.2218, | |
| "eval_samples_per_second": 0.559, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 4.216111901092501e-08, | |
| "loss": 4.3851, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "eval_loss": 4.238271236419678, | |
| "eval_runtime": 66.3755, | |
| "eval_samples_per_second": 0.557, | |
| "eval_steps_per_second": 0.557, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 0.0, | |
| "loss": 4.3848, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 4.2382659912109375, | |
| "eval_runtime": 66.5156, | |
| "eval_samples_per_second": 0.556, | |
| "eval_steps_per_second": 0.556, | |
| "step": 1900 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 13, | |
| "save_steps": 50, | |
| "total_flos": 7.73200764370944e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |