| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.561104162339198, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0716, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.387912734646695, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0333, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.9836285447878055, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0256, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.9800884023881364, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9827, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.9442251407780957, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9239, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.08419475313967, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9741, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.109925933499094, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9141, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.0673259525610956, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8904, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.3375902380289193, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8498, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6571536728673029, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8233, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.3301047327078876, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7883, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.211011357001183, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7481, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.3193538850322144, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7535, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9080855415027668, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7512, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0512591851739732, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7282, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.1161230692325639, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7459, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.7835001228910295, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7255, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.837107600828801, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7025, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7867376905857504, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7088, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6609994911085901, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6999, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.7070866332623904, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6979, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8951224417268967, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6508, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7185235221150691, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6261, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.9374694903200345, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6455, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7468011278611548, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6305, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.8806598538095017, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6133, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6587287565394545, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6315, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7936517316909375, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6329, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5556624374324638, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.615, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5645976671324228, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6014, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5406600060470248, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6091, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.48509944530499666, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5748, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5468192156383544, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5622, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.46164696458342497, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5555, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5565314990680137, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5386, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5082394365610078, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5682, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.47372540244322797, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5508, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.41615288713508297, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5449, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.45704548503040643, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.538, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4365371332944041, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5317, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4018586903717012, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5303, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.4137162417732452, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5406, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4324252417717255, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.4931, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.36079983507281077, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.496, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.39651029766952745, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5065, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.37164149719125644, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.4968, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.44175971543918185, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5167, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.43066151889582394, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.478, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.43747108772437127, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5054, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.37307042597663015, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5036, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.36725795450264026, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.481, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.35962626797899955, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4931, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.40183445913049437, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4508, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.39120770284598194, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4842, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.3068995631376283, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4532, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3090845491159116, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.45, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.34761836291617715, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4759, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.3309196033174949, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4821, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.33063742119211254, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4722, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3153825652124885, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4434, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3041050883727483, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.487, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.30797227976594754, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4655, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3315394099214905, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4747, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.304933301231908, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4624, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.28896065611928046, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4664, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.2883682511546891, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4517, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.3081527797067212, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.453, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.28225362394511627, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4606, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2691175677873696, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.466, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.27657233615825566, | |
| "learning_rate": 0.0, | |
| "loss": 0.4437, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.3958341273255936e+17, | |
| "train_loss": 0.6163317041737693, | |
| "train_runtime": 8464.3741, | |
| "train_samples_per_second": 0.827, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3958341273255936e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |