| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.0, |
| "eval_steps": 400, |
| "global_step": 7015, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07127583749109052, |
| "grad_norm": 3.4722585678100586, |
| "learning_rate": 4.8823948681397006e-05, |
| "loss": 0.4528, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.14255167498218105, |
| "grad_norm": 4.105640411376953, |
| "learning_rate": 4.76360180565455e-05, |
| "loss": 0.4275, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.21382751247327156, |
| "grad_norm": 3.594306468963623, |
| "learning_rate": 4.644808743169399e-05, |
| "loss": 0.4413, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.2851033499643621, |
| "grad_norm": 2.822234630584717, |
| "learning_rate": 4.5260156806842485e-05, |
| "loss": 0.4088, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2851033499643621, |
| "eval_bleu": 53.90945744364011, |
| "eval_chrf": 76.49653340048762, |
| "eval_loss": 1.2396413087844849, |
| "eval_runtime": 45.5827, |
| "eval_samples_per_second": 9.302, |
| "eval_steps_per_second": 1.163, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.3563791874554526, |
| "grad_norm": 1.6471904516220093, |
| "learning_rate": 4.407222618199097e-05, |
| "loss": 0.401, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4276550249465431, |
| "grad_norm": 1.9229692220687866, |
| "learning_rate": 4.288429555713946e-05, |
| "loss": 0.4401, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.4989308624376336, |
| "grad_norm": 4.6807146072387695, |
| "learning_rate": 4.1696364932287956e-05, |
| "loss": 0.458, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.5702066999287242, |
| "grad_norm": 6.216049671173096, |
| "learning_rate": 4.050843430743645e-05, |
| "loss": 0.4244, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5702066999287242, |
| "eval_bleu": 53.985067919574576, |
| "eval_chrf": 76.4778045864617, |
| "eval_loss": 1.2613844871520996, |
| "eval_runtime": 45.1467, |
| "eval_samples_per_second": 9.392, |
| "eval_steps_per_second": 1.174, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.6414825374198146, |
| "grad_norm": 2.3269271850585938, |
| "learning_rate": 3.932050368258494e-05, |
| "loss": 0.5048, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.7127583749109052, |
| "grad_norm": 1.9887986183166504, |
| "learning_rate": 3.813257305773343e-05, |
| "loss": 0.4283, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7840342124019958, |
| "grad_norm": 2.3376266956329346, |
| "learning_rate": 3.694464243288192e-05, |
| "loss": 0.4426, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.8553100498930862, |
| "grad_norm": 6.204941749572754, |
| "learning_rate": 3.5756711808030413e-05, |
| "loss": 0.496, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.8553100498930862, |
| "eval_bleu": 54.35826381811427, |
| "eval_chrf": 76.58829965762467, |
| "eval_loss": 1.2592484951019287, |
| "eval_runtime": 46.2675, |
| "eval_samples_per_second": 9.164, |
| "eval_steps_per_second": 1.146, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.9265858873841768, |
| "grad_norm": 6.887669086456299, |
| "learning_rate": 3.4568781183178906e-05, |
| "loss": 0.486, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.9978617248752673, |
| "grad_norm": 4.209754943847656, |
| "learning_rate": 3.33808505583274e-05, |
| "loss": 0.5017, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.0691375623663577, |
| "grad_norm": 0.9429085850715637, |
| "learning_rate": 3.2192919933475885e-05, |
| "loss": 0.4126, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.1404133998574484, |
| "grad_norm": 3.9917092323303223, |
| "learning_rate": 3.100498930862438e-05, |
| "loss": 0.4931, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.1404133998574484, |
| "eval_bleu": 54.607724883250675, |
| "eval_chrf": 76.84759656839454, |
| "eval_loss": 1.253143310546875, |
| "eval_runtime": 44.8458, |
| "eval_samples_per_second": 9.455, |
| "eval_steps_per_second": 1.182, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.2116892373485388, |
| "grad_norm": 4.448713779449463, |
| "learning_rate": 2.981705868377287e-05, |
| "loss": 0.4985, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.2829650748396293, |
| "grad_norm": 4.838741779327393, |
| "learning_rate": 2.862912805892136e-05, |
| "loss": 0.5203, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.3542409123307197, |
| "grad_norm": 5.198791980743408, |
| "learning_rate": 2.7441197434069853e-05, |
| "loss": 0.5296, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.4255167498218104, |
| "grad_norm": 4.123748779296875, |
| "learning_rate": 2.6253266809218342e-05, |
| "loss": 0.5622, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.4255167498218104, |
| "eval_bleu": 54.551652859378116, |
| "eval_chrf": 77.10128262017952, |
| "eval_loss": 1.2271041870117188, |
| "eval_runtime": 45.7524, |
| "eval_samples_per_second": 9.267, |
| "eval_steps_per_second": 1.158, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.4967925873129009, |
| "grad_norm": 5.228442668914795, |
| "learning_rate": 2.506533618436683e-05, |
| "loss": 0.5832, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.5680684248039913, |
| "grad_norm": 2.6589853763580322, |
| "learning_rate": 2.3877405559515324e-05, |
| "loss": 0.5352, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.639344262295082, |
| "grad_norm": 4.812443256378174, |
| "learning_rate": 2.2689474934663817e-05, |
| "loss": 0.5949, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.7106200997861725, |
| "grad_norm": 3.439000368118286, |
| "learning_rate": 2.1501544309812307e-05, |
| "loss": 0.6057, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.7106200997861725, |
| "eval_bleu": 55.315616609389814, |
| "eval_chrf": 77.2954024298945, |
| "eval_loss": 1.2040876150131226, |
| "eval_runtime": 46.0789, |
| "eval_samples_per_second": 9.202, |
| "eval_steps_per_second": 1.15, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.781895937277263, |
| "grad_norm": 7.226606369018555, |
| "learning_rate": 2.03136136849608e-05, |
| "loss": 0.6565, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.8531717747683536, |
| "grad_norm": 2.998560667037964, |
| "learning_rate": 1.912568306010929e-05, |
| "loss": 0.6565, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.924447612259444, |
| "grad_norm": 7.860637187957764, |
| "learning_rate": 1.7937752435257782e-05, |
| "loss": 0.6673, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.9957234497505345, |
| "grad_norm": 3.4744679927825928, |
| "learning_rate": 1.6749821810406275e-05, |
| "loss": 0.6896, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.9957234497505345, |
| "eval_bleu": 54.289161444971484, |
| "eval_chrf": 76.92398542441326, |
| "eval_loss": 1.1615335941314697, |
| "eval_runtime": 44.9642, |
| "eval_samples_per_second": 9.43, |
| "eval_steps_per_second": 1.179, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.066999287241625, |
| "grad_norm": 2.9694347381591797, |
| "learning_rate": 1.5561891185554764e-05, |
| "loss": 0.6495, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.1382751247327154, |
| "grad_norm": 5.613991737365723, |
| "learning_rate": 1.4373960560703257e-05, |
| "loss": 0.721, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.209550962223806, |
| "grad_norm": 4.432507514953613, |
| "learning_rate": 1.3186029935851746e-05, |
| "loss": 0.61, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.2808267997148968, |
| "grad_norm": 3.720311403274536, |
| "learning_rate": 1.1998099311000239e-05, |
| "loss": 0.6916, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.2808267997148968, |
| "eval_bleu": 54.50245782197904, |
| "eval_chrf": 76.54392547225733, |
| "eval_loss": 1.1484729051589966, |
| "eval_runtime": 46.1795, |
| "eval_samples_per_second": 9.182, |
| "eval_steps_per_second": 1.148, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.352102637205987, |
| "grad_norm": 4.093074798583984, |
| "learning_rate": 1.0810168686148728e-05, |
| "loss": 0.6521, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.4233784746970777, |
| "grad_norm": 3.274430274963379, |
| "learning_rate": 9.62223806129722e-06, |
| "loss": 0.6732, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.4946543121881684, |
| "grad_norm": 4.472115516662598, |
| "learning_rate": 8.434307436445712e-06, |
| "loss": 0.6547, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.5659301496792586, |
| "grad_norm": 6.019277095794678, |
| "learning_rate": 7.246376811594203e-06, |
| "loss": 0.6725, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.5659301496792586, |
| "eval_bleu": 54.62227679885763, |
| "eval_chrf": 76.83109982664264, |
| "eval_loss": 1.1441200971603394, |
| "eval_runtime": 45.9414, |
| "eval_samples_per_second": 9.229, |
| "eval_steps_per_second": 1.154, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.6372059871703493, |
| "grad_norm": 4.059388637542725, |
| "learning_rate": 6.0584461867426945e-06, |
| "loss": 0.6782, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.7084818246614395, |
| "grad_norm": 7.349984169006348, |
| "learning_rate": 4.870515561891186e-06, |
| "loss": 0.7001, |
| "step": 3800 |
| }, |
| { |
| "epoch": 2.77975766215253, |
| "grad_norm": 5.185818672180176, |
| "learning_rate": 3.682584937039677e-06, |
| "loss": 0.667, |
| "step": 3900 |
| }, |
| { |
| "epoch": 2.851033499643621, |
| "grad_norm": 4.289496898651123, |
| "learning_rate": 2.4946543121881682e-06, |
| "loss": 0.7136, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.851033499643621, |
| "eval_bleu": 54.4356613408887, |
| "eval_chrf": 76.90152284424269, |
| "eval_loss": 1.137866497039795, |
| "eval_runtime": 44.6351, |
| "eval_samples_per_second": 9.499, |
| "eval_steps_per_second": 1.187, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.9223093371347115, |
| "grad_norm": 2.778644323348999, |
| "learning_rate": 1.3067236873366596e-06, |
| "loss": 0.6648, |
| "step": 4100 |
| }, |
| { |
| "epoch": 2.9935851746258018, |
| "grad_norm": 6.11302375793457, |
| "learning_rate": 1.1879306248515088e-07, |
| "loss": 0.6361, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.0648610121168924, |
| "grad_norm": 4.8953704833984375, |
| "learning_rate": 1.9358517462580187e-05, |
| "loss": 0.6203, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.1361368496079827, |
| "grad_norm": 4.9622979164123535, |
| "learning_rate": 1.864575908766928e-05, |
| "loss": 0.6883, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.1361368496079827, |
| "eval_bleu": 54.4615551796715, |
| "eval_chrf": 76.75255355595692, |
| "eval_loss": 1.152636170387268, |
| "eval_runtime": 45.8574, |
| "eval_samples_per_second": 9.246, |
| "eval_steps_per_second": 1.156, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.2074126870990733, |
| "grad_norm": 5.924427509307861, |
| "learning_rate": 1.7933000712758377e-05, |
| "loss": 0.6534, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.278688524590164, |
| "grad_norm": 3.3319082260131836, |
| "learning_rate": 1.722024233784747e-05, |
| "loss": 0.6561, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.3499643620812547, |
| "grad_norm": 5.027530193328857, |
| "learning_rate": 1.6507483962936566e-05, |
| "loss": 0.6614, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.421240199572345, |
| "grad_norm": 12.282238006591797, |
| "learning_rate": 1.579472558802566e-05, |
| "loss": 0.6871, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.421240199572345, |
| "eval_bleu": 54.65121394864807, |
| "eval_chrf": 76.99318341492122, |
| "eval_loss": 1.1458663940429688, |
| "eval_runtime": 46.7948, |
| "eval_samples_per_second": 9.061, |
| "eval_steps_per_second": 1.133, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.4925160370634356, |
| "grad_norm": 4.480852127075195, |
| "learning_rate": 1.5081967213114755e-05, |
| "loss": 0.6364, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.563791874554526, |
| "grad_norm": 3.2055609226226807, |
| "learning_rate": 1.4369208838203848e-05, |
| "loss": 0.6718, |
| "step": 5000 |
| }, |
| { |
| "epoch": 3.6350677120456165, |
| "grad_norm": 3.66184663772583, |
| "learning_rate": 1.3656450463292945e-05, |
| "loss": 0.6814, |
| "step": 5100 |
| }, |
| { |
| "epoch": 3.706343549536707, |
| "grad_norm": 4.526830196380615, |
| "learning_rate": 1.294369208838204e-05, |
| "loss": 0.6839, |
| "step": 5200 |
| }, |
| { |
| "epoch": 3.706343549536707, |
| "eval_bleu": 54.43538330740564, |
| "eval_chrf": 76.83467170225417, |
| "eval_loss": 1.1387040615081787, |
| "eval_runtime": 47.0575, |
| "eval_samples_per_second": 9.01, |
| "eval_steps_per_second": 1.126, |
| "step": 5200 |
| }, |
| { |
| "epoch": 3.7776193870277974, |
| "grad_norm": 3.954864501953125, |
| "learning_rate": 1.2230933713471134e-05, |
| "loss": 0.7204, |
| "step": 5300 |
| }, |
| { |
| "epoch": 3.848895224518888, |
| "grad_norm": 4.9757280349731445, |
| "learning_rate": 1.1518175338560229e-05, |
| "loss": 0.6839, |
| "step": 5400 |
| }, |
| { |
| "epoch": 3.920171062009979, |
| "grad_norm": 4.421995639801025, |
| "learning_rate": 1.0805416963649323e-05, |
| "loss": 0.6234, |
| "step": 5500 |
| }, |
| { |
| "epoch": 3.991446899501069, |
| "grad_norm": 2.944256067276001, |
| "learning_rate": 1.0092658588738418e-05, |
| "loss": 0.7029, |
| "step": 5600 |
| }, |
| { |
| "epoch": 3.991446899501069, |
| "eval_bleu": 54.47230508466166, |
| "eval_chrf": 76.7475667588551, |
| "eval_loss": 1.143419861793518, |
| "eval_runtime": 46.7827, |
| "eval_samples_per_second": 9.063, |
| "eval_steps_per_second": 1.133, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.06272273699216, |
| "grad_norm": 4.3462114334106445, |
| "learning_rate": 9.379900213827513e-06, |
| "loss": 0.6416, |
| "step": 5700 |
| }, |
| { |
| "epoch": 4.13399857448325, |
| "grad_norm": 5.6380815505981445, |
| "learning_rate": 8.667141838916607e-06, |
| "loss": 0.6659, |
| "step": 5800 |
| }, |
| { |
| "epoch": 4.205274411974341, |
| "grad_norm": 3.9297401905059814, |
| "learning_rate": 7.954383464005702e-06, |
| "loss": 0.6263, |
| "step": 5900 |
| }, |
| { |
| "epoch": 4.276550249465431, |
| "grad_norm": 3.566382884979248, |
| "learning_rate": 7.2416250890947975e-06, |
| "loss": 0.688, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.276550249465431, |
| "eval_bleu": 54.58523394049827, |
| "eval_chrf": 76.97146289458047, |
| "eval_loss": 1.1411081552505493, |
| "eval_runtime": 47.418, |
| "eval_samples_per_second": 8.942, |
| "eval_steps_per_second": 1.118, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.3478260869565215, |
| "grad_norm": 3.055271625518799, |
| "learning_rate": 6.528866714183891e-06, |
| "loss": 0.6621, |
| "step": 6100 |
| }, |
| { |
| "epoch": 4.419101924447612, |
| "grad_norm": 5.575135231018066, |
| "learning_rate": 5.816108339272987e-06, |
| "loss": 0.6598, |
| "step": 6200 |
| }, |
| { |
| "epoch": 4.490377761938703, |
| "grad_norm": 2.131157159805298, |
| "learning_rate": 5.1033499643620815e-06, |
| "loss": 0.6608, |
| "step": 6300 |
| }, |
| { |
| "epoch": 4.5616535994297935, |
| "grad_norm": 5.105737209320068, |
| "learning_rate": 4.390591589451176e-06, |
| "loss": 0.6396, |
| "step": 6400 |
| }, |
| { |
| "epoch": 4.5616535994297935, |
| "eval_bleu": 54.33271899871316, |
| "eval_chrf": 76.81831179228314, |
| "eval_loss": 1.1372572183609009, |
| "eval_runtime": 47.3318, |
| "eval_samples_per_second": 8.958, |
| "eval_steps_per_second": 1.12, |
| "step": 6400 |
| }, |
| { |
| "epoch": 4.632929436920884, |
| "grad_norm": 4.760735988616943, |
| "learning_rate": 3.677833214540271e-06, |
| "loss": 0.6339, |
| "step": 6500 |
| }, |
| { |
| "epoch": 4.704205274411974, |
| "grad_norm": 3.7660269737243652, |
| "learning_rate": 2.9650748396293655e-06, |
| "loss": 0.63, |
| "step": 6600 |
| }, |
| { |
| "epoch": 4.775481111903065, |
| "grad_norm": 5.468283176422119, |
| "learning_rate": 2.2523164647184606e-06, |
| "loss": 0.6319, |
| "step": 6700 |
| }, |
| { |
| "epoch": 4.846756949394155, |
| "grad_norm": 11.634507179260254, |
| "learning_rate": 1.5395580898075552e-06, |
| "loss": 0.6241, |
| "step": 6800 |
| }, |
| { |
| "epoch": 4.846756949394155, |
| "eval_bleu": 54.55812484028709, |
| "eval_chrf": 76.97250388922113, |
| "eval_loss": 1.136724829673767, |
| "eval_runtime": 46.6989, |
| "eval_samples_per_second": 9.079, |
| "eval_steps_per_second": 1.135, |
| "step": 6800 |
| }, |
| { |
| "epoch": 4.918032786885246, |
| "grad_norm": 3.7019028663635254, |
| "learning_rate": 8.2679971489665e-07, |
| "loss": 0.6544, |
| "step": 6900 |
| }, |
| { |
| "epoch": 4.989308624376337, |
| "grad_norm": 3.0245063304901123, |
| "learning_rate": 1.1404133998574484e-07, |
| "loss": 0.7076, |
| "step": 7000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 7015, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 400, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.92057231015936e+16, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
