| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08465011286681716, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001128668171557562, | |
| "grad_norm": 46.31266784667969, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 11.9289, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001128668171557562, | |
| "eval_loss": 11.48071575164795, | |
| "eval_runtime": 218.574, | |
| "eval_samples_per_second": 3.418, | |
| "eval_steps_per_second": 1.711, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002257336343115124, | |
| "grad_norm": 32.02701187133789, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 9.1416, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003386004514672686, | |
| "grad_norm": 22.43347930908203, | |
| "learning_rate": 0.0002, | |
| "loss": 7.6477, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004514672686230248, | |
| "grad_norm": 24.434974670410156, | |
| "learning_rate": 0.0001999048221581858, | |
| "loss": 7.0388, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0056433408577878106, | |
| "grad_norm": 21.187652587890625, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 3.8237, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006772009029345372, | |
| "grad_norm": 18.851232528686523, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 2.9445, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007900677200902935, | |
| "grad_norm": 21.697521209716797, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 3.5231, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.009029345372460496, | |
| "grad_norm": 14.041197776794434, | |
| "learning_rate": 0.00019762960071199333, | |
| "loss": 2.2053, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.010158013544018058, | |
| "grad_norm": 15.442231178283691, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 2.3755, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011286681715575621, | |
| "grad_norm": 14.979207038879395, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 2.5331, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012415349887133182, | |
| "grad_norm": 11.316594123840332, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 2.2067, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013544018058690745, | |
| "grad_norm": 10.44298267364502, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.641, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014672686230248307, | |
| "grad_norm": 11.997466087341309, | |
| "learning_rate": 0.000190630778703665, | |
| "loss": 1.7946, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01580135440180587, | |
| "grad_norm": 12.115503311157227, | |
| "learning_rate": 0.00018870108331782217, | |
| "loss": 2.4233, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.016930022573363433, | |
| "grad_norm": 11.261882781982422, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.8223, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01805869074492099, | |
| "grad_norm": 10.030159950256348, | |
| "learning_rate": 0.0001843391445812886, | |
| "loss": 1.8842, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.019187358916478554, | |
| "grad_norm": 12.912628173828125, | |
| "learning_rate": 0.0001819152044288992, | |
| "loss": 2.4811, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.020316027088036117, | |
| "grad_norm": 14.034784317016602, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 2.0478, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02144469525959368, | |
| "grad_norm": 11.340625762939453, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 1.4553, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.022573363431151242, | |
| "grad_norm": 13.417366981506348, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 2.2667, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023702031602708805, | |
| "grad_norm": 10.488327026367188, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 1.8455, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.024830699774266364, | |
| "grad_norm": 8.669241905212402, | |
| "learning_rate": 0.00016755902076156604, | |
| "loss": 1.7471, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.025959367945823927, | |
| "grad_norm": 8.316575050354004, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 1.4701, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02708803611738149, | |
| "grad_norm": 9.43625259399414, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 1.6207, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.028216704288939052, | |
| "grad_norm": 8.611120223999023, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 1.642, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.028216704288939052, | |
| "eval_loss": 1.9240721464157104, | |
| "eval_runtime": 220.5343, | |
| "eval_samples_per_second": 3.387, | |
| "eval_steps_per_second": 1.696, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.029345372460496615, | |
| "grad_norm": 5.593110084533691, | |
| "learning_rate": 0.0001537299608346824, | |
| "loss": 1.0647, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.030474040632054177, | |
| "grad_norm": 9.355537414550781, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.5519, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03160270880361174, | |
| "grad_norm": 10.389217376708984, | |
| "learning_rate": 0.00014617486132350343, | |
| "loss": 1.8895, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0327313769751693, | |
| "grad_norm": 9.101856231689453, | |
| "learning_rate": 0.00014226182617406996, | |
| "loss": 1.7764, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.033860045146726865, | |
| "grad_norm": 9.08023738861084, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.7249, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.034988713318284424, | |
| "grad_norm": 8.803888320922852, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 1.5349, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03611738148984198, | |
| "grad_norm": 9.191908836364746, | |
| "learning_rate": 0.00013007057995042732, | |
| "loss": 1.8036, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03724604966139955, | |
| "grad_norm": 7.046676158905029, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 1.3115, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.03837471783295711, | |
| "grad_norm": 8.129071235656738, | |
| "learning_rate": 0.00012164396139381029, | |
| "loss": 1.7446, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.039503386004514675, | |
| "grad_norm": 7.32548189163208, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.2946, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.040632054176072234, | |
| "grad_norm": 8.391158103942871, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 1.6682, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0417607223476298, | |
| "grad_norm": 9.171820640563965, | |
| "learning_rate": 0.00010871557427476583, | |
| "loss": 1.6177, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04288939051918736, | |
| "grad_norm": 8.600780487060547, | |
| "learning_rate": 0.00010436193873653361, | |
| "loss": 2.033, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04401805869074492, | |
| "grad_norm": 7.288829326629639, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6004, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.045146726862302484, | |
| "grad_norm": 10.171875953674316, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 1.6914, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.046275395033860044, | |
| "grad_norm": 5.748844623565674, | |
| "learning_rate": 9.128442572523417e-05, | |
| "loss": 1.0927, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04740406320541761, | |
| "grad_norm": 7.213958263397217, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 1.536, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.04853273137697517, | |
| "grad_norm": 8.207345962524414, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 1.4197, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04966139954853273, | |
| "grad_norm": 9.604900360107422, | |
| "learning_rate": 7.835603860618972e-05, | |
| "loss": 1.743, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.050790067720090294, | |
| "grad_norm": 7.322094440460205, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 1.522, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05191873589164785, | |
| "grad_norm": 8.541365623474121, | |
| "learning_rate": 6.992942004957271e-05, | |
| "loss": 1.1983, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05304740406320542, | |
| "grad_norm": 7.738641262054443, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 1.1686, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05417607223476298, | |
| "grad_norm": 9.32544231414795, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 1.9263, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.055304740406320545, | |
| "grad_norm": 9.188823699951172, | |
| "learning_rate": 5.773817382593008e-05, | |
| "loss": 1.7528, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.056433408577878104, | |
| "grad_norm": 6.969271183013916, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 1.4092, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.056433408577878104, | |
| "eval_loss": 1.6539942026138306, | |
| "eval_runtime": 220.8995, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 1.693, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05756207674943566, | |
| "grad_norm": 7.862215995788574, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.3288, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.05869074492099323, | |
| "grad_norm": 6.700693130493164, | |
| "learning_rate": 4.6270039165317605e-05, | |
| "loss": 1.3217, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.05981941309255079, | |
| "grad_norm": 8.424921035766602, | |
| "learning_rate": 4.264235636489542e-05, | |
| "loss": 1.4351, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.060948081264108354, | |
| "grad_norm": 8.358750343322754, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 1.9325, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.062076749435665914, | |
| "grad_norm": 8.291664123535156, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.1282, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06320541760722348, | |
| "grad_norm": 6.986010551452637, | |
| "learning_rate": 3.244097923843398e-05, | |
| "loss": 1.2154, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.06433408577878104, | |
| "grad_norm": 5.823822975158691, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 1.2519, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0654627539503386, | |
| "grad_norm": 7.550477027893066, | |
| "learning_rate": 2.6272266318987603e-05, | |
| "loss": 1.4156, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.06659142212189616, | |
| "grad_norm": 6.91257381439209, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 1.4908, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.06772009029345373, | |
| "grad_norm": 9.72311782836914, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 1.7683, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06884875846501129, | |
| "grad_norm": 8.531071662902832, | |
| "learning_rate": 1.808479557110081e-05, | |
| "loss": 1.6604, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.06997742663656885, | |
| "grad_norm": 6.506885051727295, | |
| "learning_rate": 1.566085541871145e-05, | |
| "loss": 1.1158, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.07110609480812641, | |
| "grad_norm": 6.313655853271484, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 1.3392, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.07223476297968397, | |
| "grad_norm": 6.619772911071777, | |
| "learning_rate": 1.129891668217783e-05, | |
| "loss": 1.3964, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.07336343115124154, | |
| "grad_norm": 5.63805627822876, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 0.8831, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.0744920993227991, | |
| "grad_norm": 8.772890090942383, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.6972, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07562076749435666, | |
| "grad_norm": 7.003053188323975, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 1.1805, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.07674943566591422, | |
| "grad_norm": 6.298540115356445, | |
| "learning_rate": 4.628304925177318e-06, | |
| "loss": 1.3526, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.07787810383747178, | |
| "grad_norm": 7.14997673034668, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 1.5094, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.07900677200902935, | |
| "grad_norm": 7.56989860534668, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 1.6164, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08013544018058691, | |
| "grad_norm": 6.899630546569824, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 1.2452, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.08126410835214447, | |
| "grad_norm": 6.794870853424072, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 1.6971, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08239277652370203, | |
| "grad_norm": 6.301301956176758, | |
| "learning_rate": 3.805301908254455e-07, | |
| "loss": 1.5406, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.0835214446952596, | |
| "grad_norm": 6.646534442901611, | |
| "learning_rate": 9.517784181422019e-08, | |
| "loss": 1.4273, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.08465011286681716, | |
| "grad_norm": 6.380800724029541, | |
| "learning_rate": 0.0, | |
| "loss": 1.2305, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08465011286681716, | |
| "eval_loss": 1.612352728843689, | |
| "eval_runtime": 220.667, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 1.695, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.150341542313984e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |