| { | |
| "best_metric": 1.3635696172714233, | |
| "best_model_checkpoint": "./llama3/21-04-24-Weni-WeniGPT-Agents-Llama3-1.0.9-SFT_Experiment with SFT and Llama3 and updates in requirements-2_max_steps-330_batch_8_2024-04-21_ppid_9/checkpoint-90", | |
| "epoch": 4.821428571428571, | |
| "eval_steps": 30, | |
| "global_step": 270, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 0.6299236416816711, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 1.9266, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 0.4194231629371643, | |
| "learning_rate": 0.00019969365006623072, | |
| "loss": 1.5439, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 0.44722992181777954, | |
| "learning_rate": 0.00019845231970029773, | |
| "loss": 1.3488, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "eval_loss": 1.3763371706008911, | |
| "eval_runtime": 4.9516, | |
| "eval_samples_per_second": 9.29, | |
| "eval_steps_per_second": 2.423, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.4362790286540985, | |
| "learning_rate": 0.00019626873324023915, | |
| "loss": 1.3256, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 0.435635507106781, | |
| "learning_rate": 0.00019316378910323102, | |
| "loss": 1.2663, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "grad_norm": 0.5550065636634827, | |
| "learning_rate": 0.00018916720373012426, | |
| "loss": 1.1798, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "eval_loss": 1.3427196741104126, | |
| "eval_runtime": 4.9453, | |
| "eval_samples_per_second": 9.302, | |
| "eval_steps_per_second": 2.427, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.571713387966156, | |
| "learning_rate": 0.00018431722717876384, | |
| "loss": 1.0592, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.5714108943939209, | |
| "learning_rate": 0.0001786602770447513, | |
| "loss": 0.98, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "grad_norm": 0.5294617414474487, | |
| "learning_rate": 0.00017225049421328023, | |
| "loss": 0.9664, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "eval_loss": 1.3635696172714233, | |
| "eval_runtime": 4.9445, | |
| "eval_samples_per_second": 9.303, | |
| "eval_steps_per_second": 2.427, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "grad_norm": 0.8601852655410767, | |
| "learning_rate": 0.00016514922469380337, | |
| "loss": 0.9626, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9642857142857144, | |
| "grad_norm": 0.6094381213188171, | |
| "learning_rate": 0.00015742443249672827, | |
| "loss": 0.8046, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.7323458790779114, | |
| "learning_rate": 0.00014915004917131344, | |
| "loss": 0.7042, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "eval_loss": 1.3712576627731323, | |
| "eval_runtime": 4.9472, | |
| "eval_samples_per_second": 9.298, | |
| "eval_steps_per_second": 2.426, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.3214285714285716, | |
| "grad_norm": 0.7422382831573486, | |
| "learning_rate": 0.00014040526623013318, | |
| "loss": 0.7782, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.6805801391601562, | |
| "learning_rate": 0.000131273777232092, | |
| "loss": 0.6468, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.678571428571429, | |
| "grad_norm": 0.873374342918396, | |
| "learning_rate": 0.00012184297677777463, | |
| "loss": 0.6411, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.678571428571429, | |
| "eval_loss": 1.4060558080673218, | |
| "eval_runtime": 4.9458, | |
| "eval_samples_per_second": 9.301, | |
| "eval_steps_per_second": 2.426, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.9634660482406616, | |
| "learning_rate": 0.00011220312408329318, | |
| "loss": 0.7025, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.0357142857142856, | |
| "grad_norm": 0.8289043307304382, | |
| "learning_rate": 0.00010244647913780272, | |
| "loss": 0.6012, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.2142857142857144, | |
| "grad_norm": 0.5140141844749451, | |
| "learning_rate": 9.266641971224963e-05, | |
| "loss": 0.4672, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.2142857142857144, | |
| "eval_loss": 1.4097956418991089, | |
| "eval_runtime": 4.9459, | |
| "eval_samples_per_second": 9.301, | |
| "eval_steps_per_second": 2.426, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.392857142857143, | |
| "grad_norm": 0.714319109916687, | |
| "learning_rate": 8.295654767018254e-05, | |
| "loss": 0.519, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.9236059784889221, | |
| "learning_rate": 7.340979313384404e-05, | |
| "loss": 0.4759, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.971001923084259, | |
| "learning_rate": 6.411752507928642e-05, | |
| "loss": 0.504, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_loss": 1.4393504858016968, | |
| "eval_runtime": 4.9453, | |
| "eval_samples_per_second": 9.302, | |
| "eval_steps_per_second": 2.427, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.928571428571429, | |
| "grad_norm": 0.7627897262573242, | |
| "learning_rate": 5.5168676872725036e-05, | |
| "loss": 0.4943, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.107142857142857, | |
| "grad_norm": 0.7361128330230713, | |
| "learning_rate": 4.6648895117345084e-05, | |
| "loss": 0.4293, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.5763151049613953, | |
| "learning_rate": 3.863971995668062e-05, | |
| "loss": 0.369, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "eval_loss": 1.449671983718872, | |
| "eval_runtime": 4.9474, | |
| "eval_samples_per_second": 9.298, | |
| "eval_steps_per_second": 2.425, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.464285714285714, | |
| "grad_norm": 0.7767802476882935, | |
| "learning_rate": 3.121780467962353e-05, | |
| "loss": 0.409, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.642857142857143, | |
| "grad_norm": 0.835611879825592, | |
| "learning_rate": 2.445418209597632e-05, | |
| "loss": 0.3584, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.821428571428571, | |
| "grad_norm": 0.65262371301651, | |
| "learning_rate": 1.8413584703837615e-05, | |
| "loss": 0.4151, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.821428571428571, | |
| "eval_loss": 1.4748971462249756, | |
| "eval_runtime": 4.9505, | |
| "eval_samples_per_second": 9.292, | |
| "eval_steps_per_second": 2.424, | |
| "step": 270 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 330, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 90, | |
| "total_flos": 2.2169622763116954e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |