| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 19595, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.676547050476074, | |
| "learning_rate": 6.379178361826997e-06, | |
| "loss": 4.29, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 3.5399041175842285, | |
| "learning_rate": 1.2758356723653994e-05, | |
| "loss": 1.5873, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 2.1953482627868652, | |
| "learning_rate": 1.913753508548099e-05, | |
| "loss": 1.5261, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.1363420486450195, | |
| "learning_rate": 2.5516713447307987e-05, | |
| "loss": 1.4903, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.1203014850616455, | |
| "learning_rate": 3.189589180913499e-05, | |
| "loss": 1.4803, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.9569646120071411, | |
| "learning_rate": 3.827507017096198e-05, | |
| "loss": 1.4694, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.813926100730896, | |
| "learning_rate": 4.465424853278898e-05, | |
| "loss": 1.4698, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 1.609632968902588, | |
| "learning_rate": 4.974164327634601e-05, | |
| "loss": 1.4607, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 1.7478517293930054, | |
| "learning_rate": 4.814684868588926e-05, | |
| "loss": 1.4577, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.985445261001587, | |
| "learning_rate": 4.655205409543251e-05, | |
| "loss": 1.454, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.48637855052948, | |
| "learning_rate": 4.495725950497576e-05, | |
| "loss": 1.4566, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 1.9161055088043213, | |
| "learning_rate": 4.3362464914519015e-05, | |
| "loss": 1.4513, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 1.8389859199523926, | |
| "learning_rate": 4.1767670324062264e-05, | |
| "loss": 1.4545, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.7734977006912231, | |
| "learning_rate": 4.017287573360551e-05, | |
| "loss": 1.4484, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 2.1142358779907227, | |
| "learning_rate": 3.857808114314876e-05, | |
| "loss": 1.447, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 1.546547532081604, | |
| "learning_rate": 3.698328655269202e-05, | |
| "loss": 1.4457, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.6596064567565918, | |
| "learning_rate": 3.5388491962235266e-05, | |
| "loss": 1.4474, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 1.6050539016723633, | |
| "learning_rate": 3.3793697371778515e-05, | |
| "loss": 1.4426, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 1.8917399644851685, | |
| "learning_rate": 3.2198902781321764e-05, | |
| "loss": 1.445, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 1.8012959957122803, | |
| "learning_rate": 3.060410819086501e-05, | |
| "loss": 1.4465, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 1.4171031713485718, | |
| "learning_rate": 2.9009313600408265e-05, | |
| "loss": 1.4439, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 1.4994091987609863, | |
| "learning_rate": 2.7414519009951518e-05, | |
| "loss": 1.4443, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 1.6185240745544434, | |
| "learning_rate": 2.5819724419494773e-05, | |
| "loss": 1.4392, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 1.600825548171997, | |
| "learning_rate": 2.4224929829038022e-05, | |
| "loss": 1.4394, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 1.565808892250061, | |
| "learning_rate": 2.263013523858127e-05, | |
| "loss": 1.4421, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 1.489320993423462, | |
| "learning_rate": 2.103534064812452e-05, | |
| "loss": 1.4404, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 1.6581943035125732, | |
| "learning_rate": 1.9440546057667776e-05, | |
| "loss": 1.4378, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 1.7949219942092896, | |
| "learning_rate": 1.7845751467211024e-05, | |
| "loss": 1.4421, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 1.58501136302948, | |
| "learning_rate": 1.6250956876754277e-05, | |
| "loss": 1.4358, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 1.6215447187423706, | |
| "learning_rate": 1.4656162286297526e-05, | |
| "loss": 1.4411, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 1.8111215829849243, | |
| "learning_rate": 1.3061367695840776e-05, | |
| "loss": 1.4432, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 1.7637990713119507, | |
| "learning_rate": 1.1466573105384027e-05, | |
| "loss": 1.435, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 1.736876130104065, | |
| "learning_rate": 9.871778514927279e-06, | |
| "loss": 1.4429, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "grad_norm": 1.7990552186965942, | |
| "learning_rate": 8.276983924470528e-06, | |
| "loss": 1.4312, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "grad_norm": 1.466833233833313, | |
| "learning_rate": 6.682189334013779e-06, | |
| "loss": 1.4352, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "grad_norm": 2.320195198059082, | |
| "learning_rate": 5.08739474355703e-06, | |
| "loss": 1.4368, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 1.6469424962997437, | |
| "learning_rate": 3.492600153100281e-06, | |
| "loss": 1.4388, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "grad_norm": 1.8656694889068604, | |
| "learning_rate": 1.8978055626435315e-06, | |
| "loss": 1.4305, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "grad_norm": 1.6712565422058105, | |
| "learning_rate": 3.0301097218678235e-07, | |
| "loss": 1.4416, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 19595, | |
| "total_flos": 511693022656512.0, | |
| "train_loss": 1.525455296723993, | |
| "train_runtime": 771.3163, | |
| "train_samples_per_second": 203.218, | |
| "train_steps_per_second": 25.405 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 19595, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 511693022656512.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |