| { | |
| "best_global_step": 800, | |
| "best_metric": 0.5450772643089294, | |
| "best_model_checkpoint": "Llama-3.2-1B-it-Medical-LoRA/checkpoint-800", | |
| "epoch": 6.801812004530011, | |
| "eval_steps": 100, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11325028312570781, | |
| "grad_norm": 0.6456167697906494, | |
| "learning_rate": 0.00019075425790754258, | |
| "loss": 0.643, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22650056625141562, | |
| "grad_norm": 1.0792498588562012, | |
| "learning_rate": 0.0001664233576642336, | |
| "loss": 0.6465, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22650056625141562, | |
| "eval_loss": 0.5826964378356934, | |
| "eval_runtime": 394.3597, | |
| "eval_samples_per_second": 3.979, | |
| "eval_steps_per_second": 0.5, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33975084937712347, | |
| "grad_norm": 0.5628879070281982, | |
| "learning_rate": 0.0001420924574209246, | |
| "loss": 0.5815, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45300113250283125, | |
| "grad_norm": 0.5532649755477905, | |
| "learning_rate": 0.00011776155717761557, | |
| "loss": 0.5435, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45300113250283125, | |
| "eval_loss": 0.5700864791870117, | |
| "eval_runtime": 394.4221, | |
| "eval_samples_per_second": 3.978, | |
| "eval_steps_per_second": 0.499, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5662514156285391, | |
| "grad_norm": 0.7321934700012207, | |
| "learning_rate": 9.343065693430657e-05, | |
| "loss": 0.5816, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6795016987542469, | |
| "grad_norm": 0.43280959129333496, | |
| "learning_rate": 6.909975669099758e-05, | |
| "loss": 0.5611, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6795016987542469, | |
| "eval_loss": 0.5585977435112, | |
| "eval_runtime": 393.6528, | |
| "eval_samples_per_second": 3.986, | |
| "eval_steps_per_second": 0.5, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7927519818799547, | |
| "grad_norm": 0.5519748330116272, | |
| "learning_rate": 4.476885644768857e-05, | |
| "loss": 0.5964, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9060022650056625, | |
| "grad_norm": 0.43155646324157715, | |
| "learning_rate": 2.0437956204379563e-05, | |
| "loss": 0.5656, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9060022650056625, | |
| "eval_loss": 0.5522705912590027, | |
| "eval_runtime": 393.6183, | |
| "eval_samples_per_second": 3.986, | |
| "eval_steps_per_second": 0.5, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0203850509626273, | |
| "grad_norm": 0.5300919413566589, | |
| "learning_rate": 0.0001614712643678161, | |
| "loss": 0.5, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.1336353340883352, | |
| "grad_norm": 0.6625114679336548, | |
| "learning_rate": 0.00015687356321839082, | |
| "loss": 0.4964, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1336353340883352, | |
| "eval_loss": 0.5636059045791626, | |
| "eval_runtime": 394.3412, | |
| "eval_samples_per_second": 3.979, | |
| "eval_steps_per_second": 0.5, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.246885617214043, | |
| "grad_norm": 0.635853111743927, | |
| "learning_rate": 0.0001522758620689655, | |
| "loss": 0.5245, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.3601359003397508, | |
| "grad_norm": 0.6611766219139099, | |
| "learning_rate": 0.00014767816091954024, | |
| "loss": 0.5212, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3601359003397508, | |
| "eval_loss": 0.5590068101882935, | |
| "eval_runtime": 394.5384, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.4733861834654587, | |
| "grad_norm": 0.5332358479499817, | |
| "learning_rate": 0.00014308045977011496, | |
| "loss": 0.5143, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.5866364665911665, | |
| "grad_norm": 0.7008891701698303, | |
| "learning_rate": 0.00013848275862068967, | |
| "loss": 0.506, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5866364665911665, | |
| "eval_loss": 0.5510138869285583, | |
| "eval_runtime": 394.522, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.6998867497168741, | |
| "grad_norm": 0.6186133623123169, | |
| "learning_rate": 0.00013388505747126436, | |
| "loss": 0.5094, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8131370328425822, | |
| "grad_norm": 0.5614038109779358, | |
| "learning_rate": 0.00012928735632183907, | |
| "loss": 0.5112, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.8131370328425822, | |
| "eval_loss": 0.5450772643089294, | |
| "eval_runtime": 394.4384, | |
| "eval_samples_per_second": 3.978, | |
| "eval_steps_per_second": 0.499, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9263873159682898, | |
| "grad_norm": 0.5249345302581787, | |
| "learning_rate": 0.0001246896551724138, | |
| "loss": 0.5297, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.0407701019252547, | |
| "grad_norm": 0.49339523911476135, | |
| "learning_rate": 0.00012009195402298852, | |
| "loss": 0.4739, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.0407701019252547, | |
| "eval_loss": 0.5562955141067505, | |
| "eval_runtime": 394.4284, | |
| "eval_samples_per_second": 3.978, | |
| "eval_steps_per_second": 0.499, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.1540203850509627, | |
| "grad_norm": 0.6575009226799011, | |
| "learning_rate": 0.00011549425287356321, | |
| "loss": 0.4159, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.2672706681766703, | |
| "grad_norm": 0.6560823917388916, | |
| "learning_rate": 0.00011089655172413794, | |
| "loss": 0.4058, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.2672706681766703, | |
| "eval_loss": 0.563307523727417, | |
| "eval_runtime": 394.5149, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.3805209513023784, | |
| "grad_norm": 0.6496731638908386, | |
| "learning_rate": 0.00010629885057471265, | |
| "loss": 0.4114, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.493771234428086, | |
| "grad_norm": 0.7874748110771179, | |
| "learning_rate": 0.00010170114942528736, | |
| "loss": 0.3863, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.493771234428086, | |
| "eval_loss": 0.5611369609832764, | |
| "eval_runtime": 394.4816, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.607021517553794, | |
| "grad_norm": 0.6874073147773743, | |
| "learning_rate": 9.710344827586208e-05, | |
| "loss": 0.4063, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.7202718006795017, | |
| "grad_norm": 0.7157691717147827, | |
| "learning_rate": 9.250574712643678e-05, | |
| "loss": 0.4056, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.7202718006795017, | |
| "eval_loss": 0.5603105425834656, | |
| "eval_runtime": 394.4278, | |
| "eval_samples_per_second": 3.978, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.8335220838052093, | |
| "grad_norm": 0.7622489929199219, | |
| "learning_rate": 8.79080459770115e-05, | |
| "loss": 0.3907, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.9467723669309174, | |
| "grad_norm": 0.8641847968101501, | |
| "learning_rate": 8.33103448275862e-05, | |
| "loss": 0.4022, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.9467723669309174, | |
| "eval_loss": 0.5572757124900818, | |
| "eval_runtime": 394.51, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.061155152887882, | |
| "grad_norm": 0.8112807273864746, | |
| "learning_rate": 7.871264367816093e-05, | |
| "loss": 0.3592, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.17440543601359, | |
| "grad_norm": 0.9220572113990784, | |
| "learning_rate": 7.411494252873563e-05, | |
| "loss": 0.2851, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.17440543601359, | |
| "eval_loss": 0.594252347946167, | |
| "eval_runtime": 394.5109, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.287655719139298, | |
| "grad_norm": 0.6996991634368896, | |
| "learning_rate": 6.951724137931034e-05, | |
| "loss": 0.2895, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.4009060022650055, | |
| "grad_norm": 0.8084153532981873, | |
| "learning_rate": 6.491954022988506e-05, | |
| "loss": 0.3039, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.4009060022650055, | |
| "eval_loss": 0.5931687355041504, | |
| "eval_runtime": 394.5426, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.5141562853907136, | |
| "grad_norm": 0.7411876320838928, | |
| "learning_rate": 6.032183908045978e-05, | |
| "loss": 0.2939, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.627406568516421, | |
| "grad_norm": 0.774448573589325, | |
| "learning_rate": 5.5724137931034484e-05, | |
| "loss": 0.2971, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.627406568516421, | |
| "eval_loss": 0.6003153324127197, | |
| "eval_runtime": 394.6399, | |
| "eval_samples_per_second": 3.976, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.7406568516421292, | |
| "grad_norm": 0.7920798063278198, | |
| "learning_rate": 5.11264367816092e-05, | |
| "loss": 0.3123, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.853907134767837, | |
| "grad_norm": 0.9583544731140137, | |
| "learning_rate": 4.652873563218391e-05, | |
| "loss": 0.3065, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.853907134767837, | |
| "eval_loss": 0.5966437458992004, | |
| "eval_runtime": 394.4734, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.967157417893545, | |
| "grad_norm": 0.8226063847541809, | |
| "learning_rate": 4.1931034482758624e-05, | |
| "loss": 0.2923, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 4.081540203850509, | |
| "grad_norm": 0.9764239192008972, | |
| "learning_rate": 3.733333333333334e-05, | |
| "loss": 0.2475, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.081540203850509, | |
| "eval_loss": 0.6495981812477112, | |
| "eval_runtime": 394.6343, | |
| "eval_samples_per_second": 3.976, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.194790486976218, | |
| "grad_norm": 0.8083164095878601, | |
| "learning_rate": 3.273563218390805e-05, | |
| "loss": 0.219, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.308040770101925, | |
| "grad_norm": 0.7911280393600464, | |
| "learning_rate": 2.813793103448276e-05, | |
| "loss": 0.2024, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.308040770101925, | |
| "eval_loss": 0.6586623191833496, | |
| "eval_runtime": 394.662, | |
| "eval_samples_per_second": 3.976, | |
| "eval_steps_per_second": 0.499, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.421291053227633, | |
| "grad_norm": 0.9177639484405518, | |
| "learning_rate": 2.354022988505747e-05, | |
| "loss": 0.2329, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.534541336353341, | |
| "grad_norm": 1.0080626010894775, | |
| "learning_rate": 1.8942528735632184e-05, | |
| "loss": 0.2151, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.534541336353341, | |
| "eval_loss": 0.6551217436790466, | |
| "eval_runtime": 394.4836, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.647791619479049, | |
| "grad_norm": 0.9660719633102417, | |
| "learning_rate": 1.4344827586206897e-05, | |
| "loss": 0.2063, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.761041902604757, | |
| "grad_norm": 0.8095157742500305, | |
| "learning_rate": 9.747126436781609e-06, | |
| "loss": 0.2128, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.761041902604757, | |
| "eval_loss": 0.6547165513038635, | |
| "eval_runtime": 394.6263, | |
| "eval_samples_per_second": 3.976, | |
| "eval_steps_per_second": 0.499, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.874292185730464, | |
| "grad_norm": 0.8851621747016907, | |
| "learning_rate": 5.149425287356322e-06, | |
| "loss": 0.2092, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.987542468856172, | |
| "grad_norm": 0.7815496921539307, | |
| "learning_rate": 5.517241379310344e-07, | |
| "loss": 0.205, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.987542468856172, | |
| "eval_loss": 0.6554967761039734, | |
| "eval_runtime": 394.567, | |
| "eval_samples_per_second": 3.977, | |
| "eval_steps_per_second": 0.499, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.101925254813137, | |
| "grad_norm": 1.0154372453689575, | |
| "learning_rate": 5.4824991822047765e-05, | |
| "loss": 0.1804, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 5.215175537938845, | |
| "grad_norm": 1.0985440015792847, | |
| "learning_rate": 5.15538109257442e-05, | |
| "loss": 0.1807, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.215175537938845, | |
| "eval_loss": 0.689986526966095, | |
| "eval_runtime": 396.9641, | |
| "eval_samples_per_second": 3.952, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.3284258210645525, | |
| "grad_norm": 0.7789344787597656, | |
| "learning_rate": 4.828263002944063e-05, | |
| "loss": 0.1854, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 5.44167610419026, | |
| "grad_norm": 0.8624197840690613, | |
| "learning_rate": 4.501144913313706e-05, | |
| "loss": 0.1918, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.44167610419026, | |
| "eval_loss": 0.6940796375274658, | |
| "eval_runtime": 397.0537, | |
| "eval_samples_per_second": 3.952, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.554926387315969, | |
| "grad_norm": 0.8428093791007996, | |
| "learning_rate": 4.17402682368335e-05, | |
| "loss": 0.1915, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 5.668176670441676, | |
| "grad_norm": 0.935856819152832, | |
| "learning_rate": 3.846908734052994e-05, | |
| "loss": 0.2021, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.668176670441676, | |
| "eval_loss": 0.6885104775428772, | |
| "eval_runtime": 397.1251, | |
| "eval_samples_per_second": 3.951, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.781426953567384, | |
| "grad_norm": 1.1547690629959106, | |
| "learning_rate": 3.519790644422637e-05, | |
| "loss": 0.186, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 5.8946772366930915, | |
| "grad_norm": 1.0094610452651978, | |
| "learning_rate": 3.19267255479228e-05, | |
| "loss": 0.1975, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.8946772366930915, | |
| "eval_loss": 0.6890421509742737, | |
| "eval_runtime": 397.0581, | |
| "eval_samples_per_second": 3.952, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.009060022650057, | |
| "grad_norm": 0.7764310836791992, | |
| "learning_rate": 2.865554465161924e-05, | |
| "loss": 0.1983, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 6.122310305775764, | |
| "grad_norm": 1.0375205278396606, | |
| "learning_rate": 2.538436375531567e-05, | |
| "loss": 0.1299, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.122310305775764, | |
| "eval_loss": 0.7565902471542358, | |
| "eval_runtime": 397.0169, | |
| "eval_samples_per_second": 3.952, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.235560588901472, | |
| "grad_norm": 0.826404869556427, | |
| "learning_rate": 2.2113182859012105e-05, | |
| "loss": 0.1307, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 6.34881087202718, | |
| "grad_norm": 0.9645977020263672, | |
| "learning_rate": 1.884200196270854e-05, | |
| "loss": 0.1349, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.34881087202718, | |
| "eval_loss": 0.7629241943359375, | |
| "eval_runtime": 397.1134, | |
| "eval_samples_per_second": 3.951, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.462061155152888, | |
| "grad_norm": 1.2533210515975952, | |
| "learning_rate": 1.557082106640497e-05, | |
| "loss": 0.1227, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 6.575311438278596, | |
| "grad_norm": 0.7480612397193909, | |
| "learning_rate": 1.2299640170101408e-05, | |
| "loss": 0.1313, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.575311438278596, | |
| "eval_loss": 0.7648292183876038, | |
| "eval_runtime": 397.1134, | |
| "eval_samples_per_second": 3.951, | |
| "eval_steps_per_second": 0.496, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.688561721404303, | |
| "grad_norm": 0.930604875087738, | |
| "learning_rate": 9.028459273797842e-06, | |
| "loss": 0.1258, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 6.801812004530011, | |
| "grad_norm": 0.8044250011444092, | |
| "learning_rate": 5.757278377494276e-06, | |
| "loss": 0.1238, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.801812004530011, | |
| "eval_loss": 0.7659575343132019, | |
| "eval_runtime": 397.1904, | |
| "eval_samples_per_second": 3.95, | |
| "eval_steps_per_second": 0.496, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3087, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.388558543258583e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |