| { | |
| "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": 6.239875300761433, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8712, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.106489746862528, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.8685, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.756003113762325, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.8339, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.4049958181500837, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.7295, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.062660421665332, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7643, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.1675145279514005, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7436, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.721233007579608, | |
| "learning_rate": 2e-05, | |
| "loss": 0.725, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.317387550792723, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.7085, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.9735450138272617, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.6531, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.584882704110795, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.6566, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 2.7913075454860015, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.9799, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.4354574083835119, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.6211, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2793944544228928, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.5974, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.8058617225325431, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.5551, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.025703882242916, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.5703, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.8334751597829757, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.5781, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.895321360311637, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5789, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.763693176137074, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.5003, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7877642187473359, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.516, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6634701393832114, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.5202, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 1.3509081364069337, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.8414, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7048649440747311, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.4894, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.6771509972904859, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.4259, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5427512874454508, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.4196, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6522507011236754, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.4817, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6275485911222205, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.4554, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.6446053312361837, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.4436, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6567937523607981, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.4593, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5519546544621732, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.3773, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6660395039540182, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.4587, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8296318786781435, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5503, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.7156910114702046, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5145, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5289252413675042, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.3537, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.6329734049353616, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.3558, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5337151186610406, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.3664, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.636329425309888, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.3516, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6525685282611791, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.3769, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.5625475265344834, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.3602, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.49253005415049467, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.315, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5823092401835881, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.3646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4970886276016394, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.3177, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.829075186013859, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5589, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5370735537863816, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.3263, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.530579885383543, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.2842, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.47155076395794504, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.3036, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5351020663815, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.2684, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5976434511289782, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.2793, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.4297721431232722, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.2955, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4735435496330653, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.2995, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.367149841981071, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.2282, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4500744578658765, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.2599, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.6132481129164065, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4096, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.48746725576934297, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.2632, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.413132495276291, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.243, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.4227461848179275, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.2275, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.39040174405840194, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2408, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.42952065278508383, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.2317, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.4038469356951, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.2298, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.3946128014845571, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.253, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.38289133863694674, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.2488, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3680447245109583, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.2435, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.36444552701031696, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.2247, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.6449242182840127, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.3787, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.353250141234007, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.2248, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.3387586476158015, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.2117, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.3152830540253137, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.2003, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.36121912306124093, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.2568, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.3309223151347891, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.2077, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.3380130452846998, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.2361, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.30554640516273796, | |
| "learning_rate": 0.0, | |
| "loss": 0.2061, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 1.0705425667824026e+17, | |
| "train_loss": 0.4356019737465041, | |
| "train_runtime": 4769.8365, | |
| "train_samples_per_second": 1.468, | |
| "train_steps_per_second": 0.015 | |
| } | |
| ], | |
| "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": 1.0705425667824026e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |