| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.841269841269842, | |
| "eval_steps": 500, | |
| "global_step": 310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31746031746031744, | |
| "grad_norm": 2.053196907043457, | |
| "learning_rate": 0.00019948693233918952, | |
| "loss": 2.3733, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6349206349206349, | |
| "grad_norm": 4.835758209228516, | |
| "learning_rate": 0.00019795299412524945, | |
| "loss": 2.0557, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.8223971724510193, | |
| "learning_rate": 0.00019541392564000488, | |
| "loss": 1.9785, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2698412698412698, | |
| "grad_norm": 1.6451746225357056, | |
| "learning_rate": 0.00019189578116202307, | |
| "loss": 1.8827, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "grad_norm": 3.187594413757324, | |
| "learning_rate": 0.00018743466161445823, | |
| "loss": 1.8161, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 1.1520949602127075, | |
| "learning_rate": 0.00018207634412072764, | |
| "loss": 1.7753, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 560.3873291015625, | |
| "learning_rate": 0.0001758758122692791, | |
| "loss": 1.7127, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.5396825396825395, | |
| "grad_norm": 3.1525535583496094, | |
| "learning_rate": 0.00016889669190756868, | |
| "loss": 1.7557, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.8066505193710327, | |
| "learning_rate": 0.0001612105982547663, | |
| "loss": 1.6924, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.1746031746031744, | |
| "grad_norm": 6.147305011749268, | |
| "learning_rate": 0.00015289640103269625, | |
| "loss": 1.6697, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.492063492063492, | |
| "grad_norm": 1.1416895389556885, | |
| "learning_rate": 0.00014403941515576344, | |
| "loss": 1.6402, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.7278308868408203, | |
| "learning_rate": 0.00013473052528448201, | |
| "loss": 1.633, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.1269841269841265, | |
| "grad_norm": 0.981347382068634, | |
| "learning_rate": 0.00012506525322587207, | |
| "loss": 1.6499, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 27.774219512939453, | |
| "learning_rate": 0.00011514277775045768, | |
| "loss": 1.6055, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 2.1296420097351074, | |
| "learning_rate": 0.00010506491688387127, | |
| "loss": 1.5705, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.079365079365079, | |
| "grad_norm": 0.7905821800231934, | |
| "learning_rate": 9.493508311612874e-05, | |
| "loss": 1.6286, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.396825396825397, | |
| "grad_norm": 0.805335283279419, | |
| "learning_rate": 8.485722224954237e-05, | |
| "loss": 1.6359, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.7546270489692688, | |
| "learning_rate": 7.493474677412794e-05, | |
| "loss": 1.5344, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.031746031746032, | |
| "grad_norm": 0.7980037927627563, | |
| "learning_rate": 6.526947471551798e-05, | |
| "loss": 1.5468, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.349206349206349, | |
| "grad_norm": 0.7658030986785889, | |
| "learning_rate": 5.596058484423656e-05, | |
| "loss": 1.5378, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 1.7591676712036133, | |
| "learning_rate": 4.710359896730379e-05, | |
| "loss": 1.5446, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.984126984126984, | |
| "grad_norm": 0.8993221521377563, | |
| "learning_rate": 3.878940174523371e-05, | |
| "loss": 1.5279, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.301587301587301, | |
| "grad_norm": 0.7995801568031311, | |
| "learning_rate": 3.110330809243134e-05, | |
| "loss": 1.5253, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 0.8297848105430603, | |
| "learning_rate": 2.4124187730720917e-05, | |
| "loss": 1.5507, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "grad_norm": 0.7720061540603638, | |
| "learning_rate": 1.7923655879272393e-05, | |
| "loss": 1.5218, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.253968253968253, | |
| "grad_norm": 0.818590521812439, | |
| "learning_rate": 1.2565338385541792e-05, | |
| "loss": 1.5276, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.857406735420227, | |
| "learning_rate": 8.10421883797694e-06, | |
| "loss": 1.5014, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.8421258926391602, | |
| "learning_rate": 4.586074359995119e-06, | |
| "loss": 1.5059, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.206349206349206, | |
| "grad_norm": 0.7964168190956116, | |
| "learning_rate": 2.0470058747505516e-06, | |
| "loss": 1.5039, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 0.8193419575691223, | |
| "learning_rate": 5.130676608104845e-07, | |
| "loss": 1.4986, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.841269841269842, | |
| "grad_norm": 0.884679913520813, | |
| "learning_rate": 0.0, | |
| "loss": 1.5265, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.841269841269842, | |
| "step": 310, | |
| "total_flos": 1.511820317687808e+16, | |
| "train_loss": 1.6589944193440098, | |
| "train_runtime": 971.9769, | |
| "train_samples_per_second": 1.296, | |
| "train_steps_per_second": 0.319 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 310, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.511820317687808e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |