| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9960474308300395, | |
| "eval_steps": 16, | |
| "global_step": 63, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015810276679841896, | |
| "grad_norm": 3.106086254119873, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.2762, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015810276679841896, | |
| "eval_loss": 0.34248441457748413, | |
| "eval_runtime": 9.4978, | |
| "eval_samples_per_second": 11.266, | |
| "eval_steps_per_second": 2.843, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03162055335968379, | |
| "grad_norm": 0.6217677593231201, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.2623, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.04743083003952569, | |
| "grad_norm": 1.1419365406036377, | |
| "learning_rate": 6e-06, | |
| "loss": 0.3943, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.06324110671936758, | |
| "grad_norm": 0.8881447315216064, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1572, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07905138339920949, | |
| "grad_norm": 0.7020868062973022, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1578, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09486166007905138, | |
| "grad_norm": 1.2612448930740356, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.2245, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11067193675889328, | |
| "grad_norm": 1.2954490184783936, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.2101, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.12648221343873517, | |
| "grad_norm": 0.8434045314788818, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1529, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1422924901185771, | |
| "grad_norm": 0.9214808940887451, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.1522, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15810276679841898, | |
| "grad_norm": 1.3638683557510376, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3282, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 2.2487895488739014, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.472, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.18972332015810275, | |
| "grad_norm": 1.98398756980896, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.4099, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.20553359683794467, | |
| "grad_norm": 1.796846866607666, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.3184, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.22134387351778656, | |
| "grad_norm": 1.6363037824630737, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.3551, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.23715415019762845, | |
| "grad_norm": 1.927720308303833, | |
| "learning_rate": 3e-05, | |
| "loss": 0.405, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.25296442687747034, | |
| "grad_norm": 1.0266072750091553, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.2991, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.25296442687747034, | |
| "eval_loss": 0.21815715730190277, | |
| "eval_runtime": 8.042, | |
| "eval_samples_per_second": 13.305, | |
| "eval_steps_per_second": 3.357, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.26877470355731226, | |
| "grad_norm": 0.4768655002117157, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.2191, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2845849802371542, | |
| "grad_norm": 0.9923710823059082, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.384, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.30039525691699603, | |
| "grad_norm": 0.5063422322273254, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.1493, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.31620553359683795, | |
| "grad_norm": 0.5010712742805481, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0675, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33201581027667987, | |
| "grad_norm": 0.48852574825286865, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0739, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.4387320578098297, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0751, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.3455885648727417, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0516, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3794466403162055, | |
| "grad_norm": 0.29778942465782166, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0421, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3952569169960474, | |
| "grad_norm": 0.3956562578678131, | |
| "learning_rate": 5e-05, | |
| "loss": 0.057, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.41106719367588934, | |
| "grad_norm": 0.8842503428459167, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.1188, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.4268774703557312, | |
| "grad_norm": 0.9197725653648376, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.1128, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4426877470355731, | |
| "grad_norm": 0.9175456762313843, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.0628, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.45849802371541504, | |
| "grad_norm": 0.5987579822540283, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.0499, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.4743083003952569, | |
| "grad_norm": 0.8026472330093384, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0619, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4901185770750988, | |
| "grad_norm": 0.5789671540260315, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.0394, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.5059288537549407, | |
| "grad_norm": 0.7335872054100037, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.2007, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.5059288537549407, | |
| "eval_loss": 0.050875596702098846, | |
| "eval_runtime": 8.2485, | |
| "eval_samples_per_second": 12.972, | |
| "eval_steps_per_second": 3.273, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.6524657011032104, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.1999, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5375494071146245, | |
| "grad_norm": 1.1696100234985352, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.3284, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.5533596837944664, | |
| "grad_norm": 0.16617901623249054, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0075, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5691699604743083, | |
| "grad_norm": 0.34399452805519104, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.0108, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5849802371541502, | |
| "grad_norm": 0.39466115832328796, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.0283, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.6007905138339921, | |
| "grad_norm": 0.31488093733787537, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.0178, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.616600790513834, | |
| "grad_norm": 0.32184290885925293, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.011, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6324110671936759, | |
| "grad_norm": 0.395563006401062, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0096, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6482213438735178, | |
| "grad_norm": 0.22120489180088043, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.005, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.6640316205533597, | |
| "grad_norm": 0.3320792317390442, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.008, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6798418972332015, | |
| "grad_norm": 0.28633660078048706, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.005, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.4111138582229614, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.0084, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.7114624505928854, | |
| "grad_norm": 0.06865093857049942, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0019, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.05319277197122574, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.0016, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.7430830039525692, | |
| "grad_norm": 0.6792236566543579, | |
| "learning_rate": 9.4e-05, | |
| "loss": 0.1265, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.758893280632411, | |
| "grad_norm": 0.5738235712051392, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.0992, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.758893280632411, | |
| "eval_loss": 0.025439240038394928, | |
| "eval_runtime": 8.0993, | |
| "eval_samples_per_second": 13.211, | |
| "eval_steps_per_second": 3.334, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7747035573122529, | |
| "grad_norm": 0.8376194834709167, | |
| "learning_rate": 9.8e-05, | |
| "loss": 0.1525, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7905138339920948, | |
| "grad_norm": 0.18561908602714539, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0172, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8063241106719368, | |
| "grad_norm": 0.31320735812187195, | |
| "learning_rate": 9.85470908713026e-05, | |
| "loss": 0.0038, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.8221343873517787, | |
| "grad_norm": 0.4068452715873718, | |
| "learning_rate": 9.42728012826605e-05, | |
| "loss": 0.0178, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.8379446640316206, | |
| "grad_norm": 0.16684125363826752, | |
| "learning_rate": 8.742553740855506e-05, | |
| "loss": 0.0087, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.8537549407114624, | |
| "grad_norm": 0.05175252631306648, | |
| "learning_rate": 7.840323733655778e-05, | |
| "loss": 0.0011, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.018029799684882164, | |
| "learning_rate": 6.773024435212678e-05, | |
| "loss": 0.0006, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8853754940711462, | |
| "grad_norm": 0.03412799909710884, | |
| "learning_rate": 5.602683401276615e-05, | |
| "loss": 0.0009, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.9011857707509882, | |
| "grad_norm": 0.06215568631887436, | |
| "learning_rate": 4.397316598723385e-05, | |
| "loss": 0.0011, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.9169960474308301, | |
| "grad_norm": 0.05936681851744652, | |
| "learning_rate": 3.226975564787322e-05, | |
| "loss": 0.0011, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.932806324110672, | |
| "grad_norm": 0.045637015253305435, | |
| "learning_rate": 2.1596762663442218e-05, | |
| "loss": 0.0011, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.9486166007905138, | |
| "grad_norm": 0.03515447676181793, | |
| "learning_rate": 1.257446259144494e-05, | |
| "loss": 0.001, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9644268774703557, | |
| "grad_norm": 0.04813205078244209, | |
| "learning_rate": 5.727198717339511e-06, | |
| "loss": 0.0011, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.9802371541501976, | |
| "grad_norm": 0.05862165987491608, | |
| "learning_rate": 1.4529091286973995e-06, | |
| "loss": 0.0012, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.9960474308300395, | |
| "grad_norm": 0.07167425751686096, | |
| "learning_rate": 0.0, | |
| "loss": 0.0096, | |
| "step": 63 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 63, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 16, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9264114014617600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |