| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 79, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.012658227848101266, |
| "grad_norm": 4.577199459075928, |
| "learning_rate": 2.5e-06, |
| "loss": 1.5637, |
| "mean_token_accuracy": 0.7616023421287537, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.02531645569620253, |
| "grad_norm": 4.174136638641357, |
| "learning_rate": 5e-06, |
| "loss": 1.4351, |
| "mean_token_accuracy": 0.7928676009178162, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.0379746835443038, |
| "grad_norm": 4.255485534667969, |
| "learning_rate": 7.500000000000001e-06, |
| "loss": 1.3272, |
| "mean_token_accuracy": 0.80166095495224, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.05063291139240506, |
| "grad_norm": 6.9585747718811035, |
| "learning_rate": 1e-05, |
| "loss": 2.4398, |
| "mean_token_accuracy": 0.5990298986434937, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.06329113924050633, |
| "grad_norm": 4.3947625160217285, |
| "learning_rate": 1.25e-05, |
| "loss": 1.7421, |
| "mean_token_accuracy": 0.739130437374115, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.0759493670886076, |
| "grad_norm": 4.991696834564209, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 1.9926, |
| "mean_token_accuracy": 0.6711409687995911, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.08860759493670886, |
| "grad_norm": 2.500828266143799, |
| "learning_rate": 1.7500000000000002e-05, |
| "loss": 1.149, |
| "mean_token_accuracy": 0.8182706236839294, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.10126582278481013, |
| "grad_norm": 2.685185670852661, |
| "learning_rate": 2e-05, |
| "loss": 1.2567, |
| "mean_token_accuracy": 0.8011724352836609, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.11392405063291139, |
| "grad_norm": 4.4598846435546875, |
| "learning_rate": 1.9990212265199738e-05, |
| "loss": 1.8126, |
| "mean_token_accuracy": 0.668264627456665, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.12658227848101267, |
| "grad_norm": 2.375941276550293, |
| "learning_rate": 1.996086822074945e-05, |
| "loss": 0.9982, |
| "mean_token_accuracy": 0.8236443400382996, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.13924050632911392, |
| "grad_norm": 2.6599762439727783, |
| "learning_rate": 1.9912025308994146e-05, |
| "loss": 1.2161, |
| "mean_token_accuracy": 0.7948216795921326, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.1518987341772152, |
| "grad_norm": 3.1153042316436768, |
| "learning_rate": 1.9843779142227258e-05, |
| "loss": 1.3193, |
| "mean_token_accuracy": 0.7445255517959595, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.16455696202531644, |
| "grad_norm": 2.0124526023864746, |
| "learning_rate": 1.975626331552507e-05, |
| "loss": 0.6726, |
| "mean_token_accuracy": 0.8808012008666992, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.17721518987341772, |
| "grad_norm": 5.7194695472717285, |
| "learning_rate": 1.96496491452281e-05, |
| "loss": 0.4455, |
| "mean_token_accuracy": 0.9218368530273438, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.189873417721519, |
| "grad_norm": 2.4853527545928955, |
| "learning_rate": 1.9524145333581315e-05, |
| "loss": 0.8552, |
| "mean_token_accuracy": 0.84513920545578, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.20253164556962025, |
| "grad_norm": 3.448089838027954, |
| "learning_rate": 1.9379997560189677e-05, |
| "loss": 1.1599, |
| "mean_token_accuracy": 0.766257643699646, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.21518987341772153, |
| "grad_norm": 3.7161271572113037, |
| "learning_rate": 1.9217488001088784e-05, |
| "loss": 1.2298, |
| "mean_token_accuracy": 0.7517509460449219, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.22784810126582278, |
| "grad_norm": 3.3104989528656006, |
| "learning_rate": 1.903693477637204e-05, |
| "loss": 1.098, |
| "mean_token_accuracy": 0.7745019793510437, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.24050632911392406, |
| "grad_norm": 2.6112241744995117, |
| "learning_rate": 1.883869132745561e-05, |
| "loss": 0.5376, |
| "mean_token_accuracy": 0.8900830745697021, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.25316455696202533, |
| "grad_norm": 2.0473475456237793, |
| "learning_rate": 1.862314572520028e-05, |
| "loss": 0.707, |
| "mean_token_accuracy": 0.8837323188781738, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.26582278481012656, |
| "grad_norm": 1.3957444429397583, |
| "learning_rate": 1.8390719910244487e-05, |
| "loss": 0.3663, |
| "mean_token_accuracy": 0.922813892364502, |
| "step": 21 |
| }, |
| { |
| "epoch": 0.27848101265822783, |
| "grad_norm": 1.2244876623153687, |
| "learning_rate": 1.8141868867035745e-05, |
| "loss": 0.3149, |
| "mean_token_accuracy": 0.9418661594390869, |
| "step": 22 |
| }, |
| { |
| "epoch": 0.2911392405063291, |
| "grad_norm": 2.532215118408203, |
| "learning_rate": 1.7877079733177185e-05, |
| "loss": 0.7626, |
| "mean_token_accuracy": 0.847813069820404, |
| "step": 23 |
| }, |
| { |
| "epoch": 0.3037974683544304, |
| "grad_norm": 1.6833827495574951, |
| "learning_rate": 1.759687084583285e-05, |
| "loss": 0.4983, |
| "mean_token_accuracy": 0.8964338302612305, |
| "step": 24 |
| }, |
| { |
| "epoch": 0.31645569620253167, |
| "grad_norm": 1.9508320093154907, |
| "learning_rate": 1.7301790727058344e-05, |
| "loss": 0.5502, |
| "mean_token_accuracy": 0.8954567909240723, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3291139240506329, |
| "grad_norm": 2.1034395694732666, |
| "learning_rate": 1.6992417010043144e-05, |
| "loss": 0.8424, |
| "mean_token_accuracy": 0.8514899611473083, |
| "step": 26 |
| }, |
| { |
| "epoch": 0.34177215189873417, |
| "grad_norm": 1.5832319259643555, |
| "learning_rate": 1.666935530836651e-05, |
| "loss": 0.4172, |
| "mean_token_accuracy": 0.9154860973358154, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.35443037974683544, |
| "grad_norm": 1.2713000774383545, |
| "learning_rate": 1.6333238030480473e-05, |
| "loss": 0.3246, |
| "mean_token_accuracy": 0.9443087577819824, |
| "step": 28 |
| }, |
| { |
| "epoch": 0.3670886075949367, |
| "grad_norm": 3.5502662658691406, |
| "learning_rate": 1.5984723141740578e-05, |
| "loss": 0.9895, |
| "mean_token_accuracy": 0.7997010350227356, |
| "step": 29 |
| }, |
| { |
| "epoch": 0.379746835443038, |
| "grad_norm": 1.823020339012146, |
| "learning_rate": 1.562449287640781e-05, |
| "loss": 0.3899, |
| "mean_token_accuracy": 0.9311187267303467, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3924050632911392, |
| "grad_norm": 2.7523703575134277, |
| "learning_rate": 1.5253252402142989e-05, |
| "loss": 0.6343, |
| "mean_token_accuracy": 0.8607003688812256, |
| "step": 31 |
| }, |
| { |
| "epoch": 0.4050632911392405, |
| "grad_norm": 2.4160823822021484, |
| "learning_rate": 1.4871728439607967e-05, |
| "loss": 0.6876, |
| "mean_token_accuracy": 0.8515815138816833, |
| "step": 32 |
| }, |
| { |
| "epoch": 0.4177215189873418, |
| "grad_norm": 2.1896727085113525, |
| "learning_rate": 1.4480667839875786e-05, |
| "loss": 0.6721, |
| "mean_token_accuracy": 0.8881289958953857, |
| "step": 33 |
| }, |
| { |
| "epoch": 0.43037974683544306, |
| "grad_norm": 2.068215847015381, |
| "learning_rate": 1.408083612243465e-05, |
| "loss": 0.5826, |
| "mean_token_accuracy": 0.8925256729125977, |
| "step": 34 |
| }, |
| { |
| "epoch": 0.4430379746835443, |
| "grad_norm": 3.0841832160949707, |
| "learning_rate": 1.367301597664757e-05, |
| "loss": 0.7774, |
| "mean_token_accuracy": 0.826890766620636, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.45569620253164556, |
| "grad_norm": 1.8371667861938477, |
| "learning_rate": 1.3258005729601178e-05, |
| "loss": 0.6197, |
| "mean_token_accuracy": 0.8861749172210693, |
| "step": 36 |
| }, |
| { |
| "epoch": 0.46835443037974683, |
| "grad_norm": 2.582853078842163, |
| "learning_rate": 1.2836617783342968e-05, |
| "loss": 0.6051, |
| "mean_token_accuracy": 0.8715192675590515, |
| "step": 37 |
| }, |
| { |
| "epoch": 0.4810126582278481, |
| "grad_norm": 1.7430864572525024, |
| "learning_rate": 1.2409677024566145e-05, |
| "loss": 0.462, |
| "mean_token_accuracy": 0.9091353416442871, |
| "step": 38 |
| }, |
| { |
| "epoch": 0.4936708860759494, |
| "grad_norm": 1.7745352983474731, |
| "learning_rate": 1.1978019209855174e-05, |
| "loss": 0.529, |
| "mean_token_accuracy": 0.8964338302612305, |
| "step": 39 |
| }, |
| { |
| "epoch": 0.5063291139240507, |
| "grad_norm": 1.2692292928695679, |
| "learning_rate": 1.1542489329653024e-05, |
| "loss": 0.2786, |
| "mean_token_accuracy": 0.9301416873931885, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5189873417721519, |
| "grad_norm": 2.2072253227233887, |
| "learning_rate": 1.11039399541527e-05, |
| "loss": 0.5886, |
| "mean_token_accuracy": 0.8769429922103882, |
| "step": 41 |
| }, |
| { |
| "epoch": 0.5316455696202531, |
| "grad_norm": 1.068450927734375, |
| "learning_rate": 1.066322956435104e-05, |
| "loss": 0.1469, |
| "mean_token_accuracy": 0.9682462215423584, |
| "step": 42 |
| }, |
| { |
| "epoch": 0.5443037974683544, |
| "grad_norm": 1.348853588104248, |
| "learning_rate": 1.022122087153187e-05, |
| "loss": 0.285, |
| "mean_token_accuracy": 0.9306302070617676, |
| "step": 43 |
| }, |
| { |
| "epoch": 0.5569620253164557, |
| "grad_norm": 2.9932045936584473, |
| "learning_rate": 9.778779128468133e-06, |
| "loss": 1.3721, |
| "mean_token_accuracy": 0.6634098887443542, |
| "step": 44 |
| }, |
| { |
| "epoch": 0.569620253164557, |
| "grad_norm": 1.8176335096359253, |
| "learning_rate": 9.336770435648963e-06, |
| "loss": 0.4819, |
| "mean_token_accuracy": 0.9154860973358154, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5822784810126582, |
| "grad_norm": 1.915251612663269, |
| "learning_rate": 8.896060045847305e-06, |
| "loss": 0.5226, |
| "mean_token_accuracy": 0.9145090579986572, |
| "step": 46 |
| }, |
| { |
| "epoch": 0.5949367088607594, |
| "grad_norm": 0.8988140225410461, |
| "learning_rate": 8.457510670346976e-06, |
| "loss": 0.1563, |
| "mean_token_accuracy": 0.9643380641937256, |
| "step": 47 |
| }, |
| { |
| "epoch": 0.6075949367088608, |
| "grad_norm": 1.8772412538528442, |
| "learning_rate": 8.021980790144828e-06, |
| "loss": 0.3818, |
| "mean_token_accuracy": 0.9257450103759766, |
| "step": 48 |
| }, |
| { |
| "epoch": 0.620253164556962, |
| "grad_norm": 2.2602951526641846, |
| "learning_rate": 7.590322975433857e-06, |
| "loss": 0.8018, |
| "mean_token_accuracy": 0.8382999300956726, |
| "step": 49 |
| }, |
| { |
| "epoch": 0.6329113924050633, |
| "grad_norm": 2.8283371925354004, |
| "learning_rate": 7.163382216657033e-06, |
| "loss": 0.8419, |
| "mean_token_accuracy": 0.8194006085395813, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6455696202531646, |
| "grad_norm": 1.2480376958847046, |
| "learning_rate": 6.741994270398826e-06, |
| "loss": 0.2441, |
| "mean_token_accuracy": 0.9452857971191406, |
| "step": 51 |
| }, |
| { |
| "epoch": 0.6582278481012658, |
| "grad_norm": 0.8403208255767822, |
| "learning_rate": 6.326984023352435e-06, |
| "loss": 0.0997, |
| "mean_token_accuracy": 0.9824132919311523, |
| "step": 52 |
| }, |
| { |
| "epoch": 0.6708860759493671, |
| "grad_norm": 1.2160398960113525, |
| "learning_rate": 5.919163877565351e-06, |
| "loss": 0.268, |
| "mean_token_accuracy": 0.9452857971191406, |
| "step": 53 |
| }, |
| { |
| "epoch": 0.6835443037974683, |
| "grad_norm": 1.3382436037063599, |
| "learning_rate": 5.519332160124215e-06, |
| "loss": 0.2882, |
| "mean_token_accuracy": 0.9438202381134033, |
| "step": 54 |
| }, |
| { |
| "epoch": 0.6962025316455697, |
| "grad_norm": 2.3581056594848633, |
| "learning_rate": 5.128271560392037e-06, |
| "loss": 0.8954, |
| "mean_token_accuracy": 0.825817346572876, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.7088607594936709, |
| "grad_norm": 1.6703784465789795, |
| "learning_rate": 4.746747597857014e-06, |
| "loss": 0.3415, |
| "mean_token_accuracy": 0.9242794513702393, |
| "step": 56 |
| }, |
| { |
| "epoch": 0.7215189873417721, |
| "grad_norm": 2.5587410926818848, |
| "learning_rate": 4.375507123592194e-06, |
| "loss": 0.6192, |
| "mean_token_accuracy": 0.8506588339805603, |
| "step": 57 |
| }, |
| { |
| "epoch": 0.7341772151898734, |
| "grad_norm": 1.3810869455337524, |
| "learning_rate": 4.015276858259427e-06, |
| "loss": 0.2918, |
| "mean_token_accuracy": 0.9399120807647705, |
| "step": 58 |
| }, |
| { |
| "epoch": 0.7468354430379747, |
| "grad_norm": 1.2969639301300049, |
| "learning_rate": 3.6667619695195287e-06, |
| "loss": 0.2953, |
| "mean_token_accuracy": 0.9320957660675049, |
| "step": 59 |
| }, |
| { |
| "epoch": 0.759493670886076, |
| "grad_norm": 1.0896342992782593, |
| "learning_rate": 3.330644691633492e-06, |
| "loss": 0.203, |
| "mean_token_accuracy": 0.9672691822052002, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7721518987341772, |
| "grad_norm": 1.596189022064209, |
| "learning_rate": 3.00758298995686e-06, |
| "loss": 0.3213, |
| "mean_token_accuracy": 0.9369809627532959, |
| "step": 61 |
| }, |
| { |
| "epoch": 0.7848101265822784, |
| "grad_norm": 1.4405156373977661, |
| "learning_rate": 2.698209272941659e-06, |
| "loss": 0.264, |
| "mean_token_accuracy": 0.9526135921478271, |
| "step": 62 |
| }, |
| { |
| "epoch": 0.7974683544303798, |
| "grad_norm": 1.2865629196166992, |
| "learning_rate": 2.403129154167153e-06, |
| "loss": 0.2871, |
| "mean_token_accuracy": 0.9379580020904541, |
| "step": 63 |
| }, |
| { |
| "epoch": 0.810126582278481, |
| "grad_norm": 1.4740723371505737, |
| "learning_rate": 2.1229202668228197e-06, |
| "loss": 0.2372, |
| "mean_token_accuracy": 0.952125072479248, |
| "step": 64 |
| }, |
| { |
| "epoch": 0.8227848101265823, |
| "grad_norm": 2.030972957611084, |
| "learning_rate": 1.8581311329642592e-06, |
| "loss": 0.3303, |
| "mean_token_accuracy": 0.9413776397705078, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.8354430379746836, |
| "grad_norm": 2.725217580795288, |
| "learning_rate": 1.609280089755515e-06, |
| "loss": 0.8681, |
| "mean_token_accuracy": 0.8109517693519592, |
| "step": 66 |
| }, |
| { |
| "epoch": 0.8481012658227848, |
| "grad_norm": 2.7240331172943115, |
| "learning_rate": 1.3768542747997215e-06, |
| "loss": 0.7585, |
| "mean_token_accuracy": 0.832812488079071, |
| "step": 67 |
| }, |
| { |
| "epoch": 0.8607594936708861, |
| "grad_norm": 2.29013729095459, |
| "learning_rate": 1.161308672544389e-06, |
| "loss": 0.6158, |
| "mean_token_accuracy": 0.8671926259994507, |
| "step": 68 |
| }, |
| { |
| "epoch": 0.8734177215189873, |
| "grad_norm": 1.005314588546753, |
| "learning_rate": 9.630652236279626e-07, |
| "loss": 0.1511, |
| "mean_token_accuracy": 0.9706888198852539, |
| "step": 69 |
| }, |
| { |
| "epoch": 0.8860759493670886, |
| "grad_norm": 1.6021883487701416, |
| "learning_rate": 7.825119989112173e-07, |
| "loss": 0.3483, |
| "mean_token_accuracy": 0.9369809627532959, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8987341772151899, |
| "grad_norm": 1.7578600645065308, |
| "learning_rate": 6.200024398103255e-07, |
| "loss": 0.5573, |
| "mean_token_accuracy": 0.8949682712554932, |
| "step": 71 |
| }, |
| { |
| "epoch": 0.9113924050632911, |
| "grad_norm": 1.6289381980895996, |
| "learning_rate": 4.7585466641868696e-07, |
| "loss": 0.3583, |
| "mean_token_accuracy": 0.9384465217590332, |
| "step": 72 |
| }, |
| { |
| "epoch": 0.9240506329113924, |
| "grad_norm": 1.2381173372268677, |
| "learning_rate": 3.5035085477190143e-07, |
| "loss": 0.2733, |
| "mean_token_accuracy": 0.9467513561248779, |
| "step": 73 |
| }, |
| { |
| "epoch": 0.9367088607594937, |
| "grad_norm": 1.438234567642212, |
| "learning_rate": 2.4373668447493225e-07, |
| "loss": 0.3629, |
| "mean_token_accuracy": 0.9286761283874512, |
| "step": 74 |
| }, |
| { |
| "epoch": 0.9493670886075949, |
| "grad_norm": 2.719670057296753, |
| "learning_rate": 1.562208577727442e-07, |
| "loss": 0.9695, |
| "mean_token_accuracy": 0.8100649118423462, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.9620253164556962, |
| "grad_norm": 2.9035918712615967, |
| "learning_rate": 8.797469100585432e-08, |
| "loss": 0.6929, |
| "mean_token_accuracy": 0.8439716100692749, |
| "step": 76 |
| }, |
| { |
| "epoch": 0.9746835443037974, |
| "grad_norm": 2.0562000274658203, |
| "learning_rate": 3.913177925055189e-08, |
| "loss": 0.5664, |
| "mean_token_accuracy": 0.8861749172210693, |
| "step": 77 |
| }, |
| { |
| "epoch": 0.9873417721518988, |
| "grad_norm": 1.8154650926589966, |
| "learning_rate": 9.78773480026396e-09, |
| "loss": 0.4119, |
| "mean_token_accuracy": 0.9252564907073975, |
| "step": 78 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.671342134475708, |
| "learning_rate": 0.0, |
| "loss": 0.8681, |
| "mean_token_accuracy": 0.8234986662864685, |
| "step": 79 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 79, |
| "total_flos": 128980623360.0, |
| "train_loss": 0.6915170163670673, |
| "train_runtime": 913.13, |
| "train_samples_per_second": 0.087, |
| "train_steps_per_second": 0.087 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 79, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 128980623360.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
