| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.27952480782669464, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005590496156533892, | |
| "grad_norm": 2.64302659034729, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5342, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005590496156533892, | |
| "eval_loss": 3.5435373783111572, | |
| "eval_runtime": 101.8278, | |
| "eval_samples_per_second": 11.834, | |
| "eval_steps_per_second": 2.966, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011180992313067784, | |
| "grad_norm": 2.6918563842773438, | |
| "learning_rate": 0.0001, | |
| "loss": 3.4943, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.016771488469601678, | |
| "grad_norm": 2.663547992706299, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 3.4678, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02236198462613557, | |
| "grad_norm": 3.2312190532684326, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 3.2118, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.027952480782669462, | |
| "grad_norm": 2.7595348358154297, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 2.6675, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.033542976939203356, | |
| "grad_norm": 1.419068455696106, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 2.22, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.039133473095737246, | |
| "grad_norm": 0.7016229033470154, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 2.0622, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04472396925227114, | |
| "grad_norm": 0.6586772799491882, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.0141, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.050314465408805034, | |
| "grad_norm": 0.46751201152801514, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 1.9571, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.055904961565338925, | |
| "grad_norm": 0.7789000272750854, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.9574, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.061495457721872815, | |
| "grad_norm": 0.31843599677085876, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 1.9238, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.06708595387840671, | |
| "grad_norm": 0.27857884764671326, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 1.915, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0726764500349406, | |
| "grad_norm": 0.43947434425354004, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 1.9201, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0726764500349406, | |
| "eval_loss": 1.9032824039459229, | |
| "eval_runtime": 101.9948, | |
| "eval_samples_per_second": 11.814, | |
| "eval_steps_per_second": 2.961, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.07826694619147449, | |
| "grad_norm": 0.37218666076660156, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.9066, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.08385744234800839, | |
| "grad_norm": 0.41769397258758545, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 1.8943, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08944793850454227, | |
| "grad_norm": 0.4944588541984558, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 1.8743, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.09503843466107617, | |
| "grad_norm": 0.3327654004096985, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 1.8726, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.10062893081761007, | |
| "grad_norm": 0.4804844856262207, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.8517, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10621942697414395, | |
| "grad_norm": 0.5033496022224426, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 1.8815, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.11180992313067785, | |
| "grad_norm": 0.4059678316116333, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.8633, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11740041928721175, | |
| "grad_norm": 0.5624929070472717, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 1.8528, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.12299091544374563, | |
| "grad_norm": 0.30437928438186646, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 1.8548, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.12858141160027953, | |
| "grad_norm": 0.4757891297340393, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 1.8662, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.13417190775681342, | |
| "grad_norm": 0.5576551556587219, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 1.8525, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.13976240391334732, | |
| "grad_norm": 0.301616907119751, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 1.8448, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1453529000698812, | |
| "grad_norm": 0.3822285532951355, | |
| "learning_rate": 5e-05, | |
| "loss": 1.841, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1453529000698812, | |
| "eval_loss": 1.8403270244598389, | |
| "eval_runtime": 101.9955, | |
| "eval_samples_per_second": 11.814, | |
| "eval_steps_per_second": 2.961, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": 0.4914432764053345, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 1.8609, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15653389238294899, | |
| "grad_norm": 0.35133427381515503, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 1.8361, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.16212438853948288, | |
| "grad_norm": 0.3233548104763031, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 1.8291, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.16771488469601678, | |
| "grad_norm": 0.4606425166130066, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.855, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17330538085255065, | |
| "grad_norm": 0.7441600561141968, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 1.8387, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.17889587700908455, | |
| "grad_norm": 0.3128848373889923, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.8219, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.18448637316561844, | |
| "grad_norm": 0.33925583958625793, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 1.8055, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.19007686932215234, | |
| "grad_norm": 0.6340675950050354, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 1.8227, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.19566736547868624, | |
| "grad_norm": 0.7749468684196472, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 1.8204, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.20125786163522014, | |
| "grad_norm": 0.3163313865661621, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 1.7955, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.206848357791754, | |
| "grad_norm": 0.3138035833835602, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 1.8125, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2124388539482879, | |
| "grad_norm": 0.3796585500240326, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.7958, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.2180293501048218, | |
| "grad_norm": 0.34551334381103516, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 1.7706, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2180293501048218, | |
| "eval_loss": 1.7951258420944214, | |
| "eval_runtime": 101.982, | |
| "eval_samples_per_second": 11.816, | |
| "eval_steps_per_second": 2.961, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2236198462613557, | |
| "grad_norm": 0.5342937707901001, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.8122, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2292103424178896, | |
| "grad_norm": 0.5158257484436035, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 1.7735, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.2348008385744235, | |
| "grad_norm": 0.4467351734638214, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 1.794, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.24039133473095736, | |
| "grad_norm": 0.5328588485717773, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 1.7981, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.24598183088749126, | |
| "grad_norm": 0.30072852969169617, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.8019, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.25157232704402516, | |
| "grad_norm": 0.33151668310165405, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 1.7739, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.25716282320055905, | |
| "grad_norm": 0.5027835965156555, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 1.792, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.26275331935709295, | |
| "grad_norm": 0.2943630516529083, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 1.7813, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.26834381551362685, | |
| "grad_norm": 0.37901201844215393, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 1.8137, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.27393431167016075, | |
| "grad_norm": 0.6345136165618896, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 1.8183, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.27952480782669464, | |
| "grad_norm": 0.42185863852500916, | |
| "learning_rate": 0.0, | |
| "loss": 1.7685, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0455434309664768e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |