| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.9011406844106464, |
| "eval_steps": 50, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.6337016224861145, |
| "learning_rate": 7.5e-07, |
| "loss": 1.326, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_loss": 1.3350285291671753, |
| "eval_runtime": 394.6138, |
| "eval_samples_per_second": 5.459, |
| "eval_steps_per_second": 0.684, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.19, |
| "grad_norm": 0.6025093197822571, |
| "learning_rate": 1.5e-06, |
| "loss": 1.3065, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.19, |
| "eval_loss": 1.2850024700164795, |
| "eval_runtime": 394.5243, |
| "eval_samples_per_second": 5.46, |
| "eval_steps_per_second": 0.684, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.29, |
| "grad_norm": 0.5091537237167358, |
| "learning_rate": 2.25e-06, |
| "loss": 1.2129, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.29, |
| "eval_loss": 1.1226935386657715, |
| "eval_runtime": 394.4733, |
| "eval_samples_per_second": 5.46, |
| "eval_steps_per_second": 0.684, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.38, |
| "grad_norm": 0.6157342195510864, |
| "learning_rate": 3e-06, |
| "loss": 0.9817, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.38, |
| "eval_loss": 0.8010650873184204, |
| "eval_runtime": 394.6158, |
| "eval_samples_per_second": 5.458, |
| "eval_steps_per_second": 0.684, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.48, |
| "grad_norm": 0.22548706829547882, |
| "learning_rate": 2.971177920604846e-06, |
| "loss": 0.6524, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_loss": 0.5880268812179565, |
| "eval_runtime": 394.5715, |
| "eval_samples_per_second": 5.459, |
| "eval_steps_per_second": 0.684, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.57, |
| "grad_norm": 0.19589273631572723, |
| "learning_rate": 2.88581929876693e-06, |
| "loss": 0.5671, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.57, |
| "eval_loss": 0.563366174697876, |
| "eval_runtime": 394.434, |
| "eval_samples_per_second": 5.461, |
| "eval_steps_per_second": 0.685, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.67, |
| "grad_norm": 0.20705421268939972, |
| "learning_rate": 2.747204418453818e-06, |
| "loss": 0.5519, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.67, |
| "eval_loss": 0.547275722026825, |
| "eval_runtime": 394.6461, |
| "eval_samples_per_second": 5.458, |
| "eval_steps_per_second": 0.684, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.76, |
| "grad_norm": 0.1976100355386734, |
| "learning_rate": 2.5606601717798212e-06, |
| "loss": 0.5389, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.76, |
| "eval_loss": 0.5334580540657043, |
| "eval_runtime": 394.6493, |
| "eval_samples_per_second": 5.458, |
| "eval_steps_per_second": 0.684, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.86, |
| "grad_norm": 0.18715547025203705, |
| "learning_rate": 2.3333553495294033e-06, |
| "loss": 0.5297, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.86, |
| "eval_loss": 0.5227681994438171, |
| "eval_runtime": 394.7112, |
| "eval_samples_per_second": 5.457, |
| "eval_steps_per_second": 0.684, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.95, |
| "grad_norm": 0.1926855742931366, |
| "learning_rate": 2.074025148547635e-06, |
| "loss": 0.5114, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.95, |
| "eval_loss": 0.5151167511940002, |
| "eval_runtime": 390.7412, |
| "eval_samples_per_second": 5.513, |
| "eval_steps_per_second": 0.691, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.05, |
| "grad_norm": 0.20605137944221497, |
| "learning_rate": 1.7926354830241926e-06, |
| "loss": 0.5109, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.05, |
| "eval_loss": 0.5088621973991394, |
| "eval_runtime": 392.8267, |
| "eval_samples_per_second": 5.483, |
| "eval_steps_per_second": 0.687, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.14, |
| "grad_norm": 0.23452071845531464, |
| "learning_rate": 1.5e-06, |
| "loss": 0.5079, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.14, |
| "eval_loss": 0.5038764476776123, |
| "eval_runtime": 391.3908, |
| "eval_samples_per_second": 5.503, |
| "eval_steps_per_second": 0.69, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.24, |
| "grad_norm": 0.22774113714694977, |
| "learning_rate": 1.2073645169758077e-06, |
| "loss": 0.4999, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_loss": 0.5000301599502563, |
| "eval_runtime": 392.1009, |
| "eval_samples_per_second": 5.493, |
| "eval_steps_per_second": 0.689, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.33, |
| "grad_norm": 0.21539181470870972, |
| "learning_rate": 9.259748514523654e-07, |
| "loss": 0.4908, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.33, |
| "eval_loss": 0.4972001016139984, |
| "eval_runtime": 392.3879, |
| "eval_samples_per_second": 5.489, |
| "eval_steps_per_second": 0.688, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.43, |
| "grad_norm": 0.20129938423633575, |
| "learning_rate": 6.666446504705971e-07, |
| "loss": 0.4888, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.43, |
| "eval_loss": 0.4950333833694458, |
| "eval_runtime": 392.4542, |
| "eval_samples_per_second": 5.489, |
| "eval_steps_per_second": 0.688, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.52, |
| "grad_norm": 0.20948025584220886, |
| "learning_rate": 4.3933982822017883e-07, |
| "loss": 0.4915, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_loss": 0.49351951479911804, |
| "eval_runtime": 392.3544, |
| "eval_samples_per_second": 5.49, |
| "eval_steps_per_second": 0.688, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.62, |
| "grad_norm": 0.21380692720413208, |
| "learning_rate": 2.52795581546182e-07, |
| "loss": 0.4935, |
| "step": 850 |
| }, |
| { |
| "epoch": 1.62, |
| "eval_loss": 0.4924446642398834, |
| "eval_runtime": 392.3544, |
| "eval_samples_per_second": 5.49, |
| "eval_steps_per_second": 0.688, |
| "step": 850 |
| }, |
| { |
| "epoch": 1.71, |
| "grad_norm": 0.2027772217988968, |
| "learning_rate": 1.141807012330699e-07, |
| "loss": 0.4869, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.71, |
| "eval_loss": 0.4918237328529358, |
| "eval_runtime": 392.4738, |
| "eval_samples_per_second": 5.488, |
| "eval_steps_per_second": 0.688, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.81, |
| "grad_norm": 0.22568389773368835, |
| "learning_rate": 2.8822079395154353e-08, |
| "loss": 0.4887, |
| "step": 950 |
| }, |
| { |
| "epoch": 1.81, |
| "eval_loss": 0.4916115403175354, |
| "eval_runtime": 392.4881, |
| "eval_samples_per_second": 5.488, |
| "eval_steps_per_second": 0.688, |
| "step": 950 |
| }, |
| { |
| "epoch": 1.9, |
| "grad_norm": 0.2142002135515213, |
| "learning_rate": 0.0, |
| "loss": 0.4898, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.9, |
| "eval_loss": 0.4915645122528076, |
| "eval_runtime": 392.486, |
| "eval_samples_per_second": 5.488, |
| "eval_steps_per_second": 0.688, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 200, |
| "total_flos": 2.664713513598517e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
