| { | |
| "best_metric": 1.9750508069992065, | |
| "best_model_checkpoint": "./gemma-python/checkpoint-40", | |
| "epoch": 10.0, | |
| "eval_steps": 2, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 40.636978402335416, | |
| "learning_rate": 0.0001, | |
| "loss": 19.0016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 18.6992130279541, | |
| "eval_runtime": 2.881, | |
| "eval_samples_per_second": 7.289, | |
| "eval_steps_per_second": 1.041, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 41.61053527062362, | |
| "learning_rate": 0.0002, | |
| "loss": 19.4686, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 16.257802963256836, | |
| "eval_runtime": 2.9111, | |
| "eval_samples_per_second": 7.214, | |
| "eval_steps_per_second": 1.031, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 28.704819713850974, | |
| "learning_rate": 0.00019991889981715698, | |
| "loss": 13.2303, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 26.40444243073739, | |
| "learning_rate": 0.00019967573081342103, | |
| "loss": 11.468, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 8.28911018371582, | |
| "eval_runtime": 2.9257, | |
| "eval_samples_per_second": 7.178, | |
| "eval_steps_per_second": 1.025, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 12.912981323843146, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 9.3107, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 7.943058500648636, | |
| "learning_rate": 0.00019870502626379127, | |
| "loss": 7.5305, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 5.884701728820801, | |
| "eval_runtime": 2.9479, | |
| "eval_samples_per_second": 7.124, | |
| "eval_steps_per_second": 1.018, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 6.267657551985817, | |
| "learning_rate": 0.00019797906520422677, | |
| "loss": 6.6492, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.0825555341832365, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 5.7572, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 4.363473892211914, | |
| "eval_runtime": 2.9653, | |
| "eval_samples_per_second": 7.082, | |
| "eval_steps_per_second": 1.012, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 4.88565620317727, | |
| "learning_rate": 0.00019605181116313724, | |
| "loss": 4.5414, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 5.0847008955317605, | |
| "learning_rate": 0.00019485364419471454, | |
| "loss": 4.3903, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 3.284867763519287, | |
| "eval_runtime": 2.9746, | |
| "eval_samples_per_second": 7.06, | |
| "eval_steps_per_second": 1.009, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 3.424587898800574, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 3.79, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.7255824385278506, | |
| "learning_rate": 0.00019199794436588243, | |
| "loss": 2.9497, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 2.853942394256592, | |
| "eval_runtime": 2.9866, | |
| "eval_samples_per_second": 7.031, | |
| "eval_steps_per_second": 1.004, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 2.1001906898750624, | |
| "learning_rate": 0.00019034504346103823, | |
| "loss": 2.7728, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.9200021565941778, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 2.8738, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 2.62028431892395, | |
| "eval_runtime": 2.9982, | |
| "eval_samples_per_second": 7.004, | |
| "eval_steps_per_second": 1.001, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.8837224890225774, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 3.0787, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.8929687978608318, | |
| "learning_rate": 0.0001845190085543795, | |
| "loss": 2.7298, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.453444242477417, | |
| "eval_runtime": 2.9964, | |
| "eval_samples_per_second": 7.008, | |
| "eval_steps_per_second": 1.001, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 1.3652069569291694, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 2.5967, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 2.4263600812149417, | |
| "learning_rate": 0.00017994427634035015, | |
| "loss": 2.4284, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 2.307706832885742, | |
| "eval_runtime": 2.9963, | |
| "eval_samples_per_second": 7.009, | |
| "eval_steps_per_second": 1.001, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 2.5673391658400053, | |
| "learning_rate": 0.00017746049618276545, | |
| "loss": 2.6721, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.2252437500899656, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 2.394, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 2.187636137008667, | |
| "eval_runtime": 2.9975, | |
| "eval_samples_per_second": 7.006, | |
| "eval_steps_per_second": 1.001, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 2.345233295279928, | |
| "learning_rate": 0.00017212024473438147, | |
| "loss": 2.3972, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 1.1122620317353238, | |
| "learning_rate": 0.00016927243535095997, | |
| "loss": 2.069, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 2.1294100284576416, | |
| "eval_runtime": 2.993, | |
| "eval_samples_per_second": 7.016, | |
| "eval_steps_per_second": 1.002, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 2.8270209249093803, | |
| "learning_rate": 0.00016631226582407952, | |
| "loss": 2.211, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.323169716541166, | |
| "learning_rate": 0.00016324453755953773, | |
| "loss": 1.9355, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.1047682762145996, | |
| "eval_runtime": 2.9871, | |
| "eval_samples_per_second": 7.03, | |
| "eval_steps_per_second": 1.004, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 1.9938311808450486, | |
| "learning_rate": 0.0001600742264237979, | |
| "loss": 2.1962, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 3.330986691029466, | |
| "learning_rate": 0.00015680647467311557, | |
| "loss": 1.9635, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_loss": 2.0707101821899414, | |
| "eval_runtime": 2.9895, | |
| "eval_samples_per_second": 7.025, | |
| "eval_steps_per_second": 1.004, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "grad_norm": 2.0371854480792178, | |
| "learning_rate": 0.0001534465826127801, | |
| "loss": 2.2319, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 3.2163831286077653, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.092, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 2.059619426727295, | |
| "eval_runtime": 2.9996, | |
| "eval_samples_per_second": 7.001, | |
| "eval_steps_per_second": 1.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "grad_norm": 2.853987323853131, | |
| "learning_rate": 0.00014647231720437686, | |
| "loss": 1.9182, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 2.2997509863024352, | |
| "learning_rate": 0.00014286925614030542, | |
| "loss": 1.9675, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_loss": 2.0287458896636963, | |
| "eval_runtime": 2.9966, | |
| "eval_samples_per_second": 7.008, | |
| "eval_steps_per_second": 1.001, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "grad_norm": 2.2770679758385244, | |
| "learning_rate": 0.00013919666098600753, | |
| "loss": 1.9815, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.8553765652252152, | |
| "learning_rate": 0.00013546048870425356, | |
| "loss": 1.9693, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.022012710571289, | |
| "eval_runtime": 2.9895, | |
| "eval_samples_per_second": 7.025, | |
| "eval_steps_per_second": 1.004, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "grad_norm": 3.8094922067262336, | |
| "learning_rate": 0.00013166679938014726, | |
| "loss": 1.6479, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 3.5435911597121277, | |
| "learning_rate": 0.0001278217463916453, | |
| "loss": 2.0198, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_loss": 2.012432336807251, | |
| "eval_runtime": 2.9987, | |
| "eval_samples_per_second": 7.003, | |
| "eval_steps_per_second": 1.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 1.4676241516417539, | |
| "learning_rate": 0.0001239315664287558, | |
| "loss": 1.7496, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 1.4772602834377506, | |
| "learning_rate": 0.00012000256937760445, | |
| "loss": 1.9357, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 1.9945744276046753, | |
| "eval_runtime": 3.0019, | |
| "eval_samples_per_second": 6.995, | |
| "eval_steps_per_second": 0.999, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 0.8198622785029981, | |
| "learning_rate": 0.00011604112808577603, | |
| "loss": 1.8365, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 2.5267989029749556, | |
| "learning_rate": 0.0001120536680255323, | |
| "loss": 1.8147, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_loss": 1.9979486465454102, | |
| "eval_runtime": 2.9865, | |
| "eval_samples_per_second": 7.032, | |
| "eval_steps_per_second": 1.005, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 1.2889515222114942, | |
| "learning_rate": 0.00010804665687167262, | |
| "loss": 1.6703, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.3474067788797102, | |
| "learning_rate": 0.00010402659401094152, | |
| "loss": 1.9084, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.9750508069992065, | |
| "eval_runtime": 2.9945, | |
| "eval_samples_per_second": 7.013, | |
| "eval_steps_per_second": 1.002, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 1.320063776368443, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6233, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 0.7858628087737163, | |
| "learning_rate": 9.597340598905852e-05, | |
| "loss": 1.6678, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "eval_loss": 2.004897356033325, | |
| "eval_runtime": 2.9946, | |
| "eval_samples_per_second": 7.013, | |
| "eval_steps_per_second": 1.002, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "grad_norm": 1.149181462350102, | |
| "learning_rate": 9.195334312832742e-05, | |
| "loss": 1.5673, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 1.961547695831496, | |
| "learning_rate": 8.79463319744677e-05, | |
| "loss": 1.7639, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 1.9885122776031494, | |
| "eval_runtime": 2.9905, | |
| "eval_samples_per_second": 7.022, | |
| "eval_steps_per_second": 1.003, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "grad_norm": 0.794217334050356, | |
| "learning_rate": 8.395887191422397e-05, | |
| "loss": 1.6191, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 1.5568588659062292, | |
| "learning_rate": 7.999743062239557e-05, | |
| "loss": 1.7475, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "eval_loss": 1.9777300357818604, | |
| "eval_runtime": 2.9821, | |
| "eval_samples_per_second": 7.042, | |
| "eval_steps_per_second": 1.006, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 0.9110203190054421, | |
| "learning_rate": 7.606843357124426e-05, | |
| "loss": 1.5998, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.4501990937976796, | |
| "learning_rate": 7.217825360835473e-05, | |
| "loss": 1.4848, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.9939006567001343, | |
| "eval_runtime": 2.9785, | |
| "eval_samples_per_second": 7.05, | |
| "eval_steps_per_second": 1.007, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "grad_norm": 1.3413384555399062, | |
| "learning_rate": 6.833320061985277e-05, | |
| "loss": 1.5343, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.9844954583473513, | |
| "learning_rate": 6.453951129574644e-05, | |
| "loss": 1.3065, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_loss": 2.0264320373535156, | |
| "eval_runtime": 2.9839, | |
| "eval_samples_per_second": 7.038, | |
| "eval_steps_per_second": 1.005, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "grad_norm": 1.268663878876962, | |
| "learning_rate": 6.080333901399251e-05, | |
| "loss": 1.4153, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 1.1638516740810099, | |
| "learning_rate": 5.713074385969457e-05, | |
| "loss": 1.4792, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "eval_loss": 2.012540817260742, | |
| "eval_runtime": 2.9954, | |
| "eval_samples_per_second": 7.011, | |
| "eval_steps_per_second": 1.002, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "grad_norm": 0.8956974540095054, | |
| "learning_rate": 5.3527682795623146e-05, | |
| "loss": 1.5184, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.8166104294104601, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.4233, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "eval_loss": 2.0203704833984375, | |
| "eval_runtime": 2.9966, | |
| "eval_samples_per_second": 7.008, | |
| "eval_steps_per_second": 1.001, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 1.2567309830006292, | |
| "learning_rate": 4.6553417387219886e-05, | |
| "loss": 1.5766, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.202021898168564, | |
| "learning_rate": 4.3193525326884435e-05, | |
| "loss": 1.2534, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.0317745208740234, | |
| "eval_runtime": 2.9887, | |
| "eval_samples_per_second": 7.027, | |
| "eval_steps_per_second": 1.004, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "grad_norm": 1.0179404054971375, | |
| "learning_rate": 3.99257735762021e-05, | |
| "loss": 1.3538, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 0.8024465225797554, | |
| "learning_rate": 3.675546244046228e-05, | |
| "loss": 1.2409, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_loss": 2.0444860458374023, | |
| "eval_runtime": 2.9957, | |
| "eval_samples_per_second": 7.01, | |
| "eval_steps_per_second": 1.001, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "grad_norm": 1.0938821440297672, | |
| "learning_rate": 3.36877341759205e-05, | |
| "loss": 1.2446, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 1.4397725924431397, | |
| "learning_rate": 3.072756464904006e-05, | |
| "loss": 1.4309, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_loss": 2.0641307830810547, | |
| "eval_runtime": 3.0002, | |
| "eval_samples_per_second": 6.999, | |
| "eval_steps_per_second": 1.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "grad_norm": 1.084317322881849, | |
| "learning_rate": 2.7879755265618555e-05, | |
| "loss": 1.4057, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 0.8921847488708302, | |
| "learning_rate": 2.514892518288988e-05, | |
| "loss": 1.1622, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "eval_loss": 2.0632762908935547, | |
| "eval_runtime": 2.9934, | |
| "eval_samples_per_second": 7.015, | |
| "eval_steps_per_second": 1.002, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "grad_norm": 1.2733235220422945, | |
| "learning_rate": 2.2539503817234553e-05, | |
| "loss": 1.2667, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.01591405423162, | |
| "learning_rate": 2.0055723659649904e-05, | |
| "loss": 1.228, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.09301495552063, | |
| "eval_runtime": 2.9938, | |
| "eval_samples_per_second": 7.014, | |
| "eval_steps_per_second": 1.002, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 0.9494450303367244, | |
| "learning_rate": 1.7701613410634365e-05, | |
| "loss": 1.1147, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 0.8254286577206483, | |
| "learning_rate": 1.5480991445620542e-05, | |
| "loss": 1.3076, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "eval_loss": 2.1076860427856445, | |
| "eval_runtime": 2.9974, | |
| "eval_samples_per_second": 7.006, | |
| "eval_steps_per_second": 1.001, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "grad_norm": 0.9874923331530434, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 1.1572, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.8701092754993289, | |
| "learning_rate": 1.1454397434679021e-05, | |
| "loss": 1.2323, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "eval_loss": 2.1060104370117188, | |
| "eval_runtime": 2.9923, | |
| "eval_samples_per_second": 7.018, | |
| "eval_steps_per_second": 1.003, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "grad_norm": 0.9048894666644874, | |
| "learning_rate": 9.65495653896179e-06, | |
| "loss": 1.1888, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.8899151834513122, | |
| "learning_rate": 8.002055634117578e-06, | |
| "loss": 1.1635, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "eval_loss": 2.1039013862609863, | |
| "eval_runtime": 2.9883, | |
| "eval_samples_per_second": 7.027, | |
| "eval_steps_per_second": 1.004, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "grad_norm": 0.9759646607551775, | |
| "learning_rate": 6.498375731458528e-06, | |
| "loss": 1.0924, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.067529387326401, | |
| "learning_rate": 5.146355805285452e-06, | |
| "loss": 1.261, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.1068060398101807, | |
| "eval_runtime": 2.9995, | |
| "eval_samples_per_second": 7.001, | |
| "eval_steps_per_second": 1.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "grad_norm": 0.6912625265797223, | |
| "learning_rate": 3.948188836862776e-06, | |
| "loss": 1.2225, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.8624797224342494, | |
| "learning_rate": 2.905818257394799e-06, | |
| "loss": 1.0122, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "eval_loss": 2.1110289096832275, | |
| "eval_runtime": 2.9973, | |
| "eval_samples_per_second": 7.006, | |
| "eval_steps_per_second": 1.001, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "grad_norm": 0.7882237650664056, | |
| "learning_rate": 2.0209347957732328e-06, | |
| "loss": 1.0959, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 0.8572353081855683, | |
| "learning_rate": 1.2949737362087156e-06, | |
| "loss": 1.218, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "eval_loss": 2.117999315261841, | |
| "eval_runtime": 2.9874, | |
| "eval_samples_per_second": 7.03, | |
| "eval_steps_per_second": 1.004, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "grad_norm": 0.8712624014542376, | |
| "learning_rate": 7.291125901946027e-07, | |
| "loss": 1.2579, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 0.7643303727279644, | |
| "learning_rate": 3.2426918657900704e-07, | |
| "loss": 1.1022, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "eval_loss": 2.1226046085357666, | |
| "eval_runtime": 2.9845, | |
| "eval_samples_per_second": 7.036, | |
| "eval_steps_per_second": 1.005, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "grad_norm": 0.7335554379014946, | |
| "learning_rate": 8.110018284304133e-08, | |
| "loss": 1.1355, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.7141910762762422, | |
| "learning_rate": 0.0, | |
| "loss": 1.2072, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.1142799854278564, | |
| "eval_runtime": 2.9807, | |
| "eval_samples_per_second": 7.045, | |
| "eval_steps_per_second": 1.006, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 8, | |
| "total_flos": 3.704687606680781e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |