| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.5, | |
| "eval_steps": 1000, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 6.840716361999512, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 3.0263, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 3.3429949283599854, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 1.6744, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 5.441232681274414, | |
| "learning_rate": 5.96e-05, | |
| "loss": 1.5933, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 2.98654842376709, | |
| "learning_rate": 7.960000000000001e-05, | |
| "loss": 1.5397, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.9203338623046875, | |
| "learning_rate": 9.960000000000001e-05, | |
| "loss": 1.5443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.706545352935791, | |
| "learning_rate": 9.966779661016949e-05, | |
| "loss": 1.5593, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.7645795345306396, | |
| "learning_rate": 9.932881355932204e-05, | |
| "loss": 1.569, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.51504647731781, | |
| "learning_rate": 9.898983050847458e-05, | |
| "loss": 1.5094, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.9701757431030273, | |
| "learning_rate": 9.865084745762713e-05, | |
| "loss": 1.5007, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.04455304145813, | |
| "learning_rate": 9.831186440677966e-05, | |
| "loss": 1.5154, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 1.4852064847946167, | |
| "eval_runtime": 93.0778, | |
| "eval_samples_per_second": 10.744, | |
| "eval_steps_per_second": 2.686, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.8170721530914307, | |
| "learning_rate": 9.797288135593221e-05, | |
| "loss": 1.4842, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.3413264751434326, | |
| "learning_rate": 9.763389830508474e-05, | |
| "loss": 1.4884, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.753246784210205, | |
| "learning_rate": 9.72949152542373e-05, | |
| "loss": 1.464, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 3.4083104133605957, | |
| "learning_rate": 9.695593220338983e-05, | |
| "loss": 1.4739, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 2.3532192707061768, | |
| "learning_rate": 9.661694915254238e-05, | |
| "loss": 1.5031, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.6274845600128174, | |
| "learning_rate": 9.627796610169491e-05, | |
| "loss": 1.4936, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.692795991897583, | |
| "learning_rate": 9.593898305084746e-05, | |
| "loss": 1.4842, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 2.3271782398223877, | |
| "learning_rate": 9.56e-05, | |
| "loss": 1.527, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.4074125289916992, | |
| "learning_rate": 9.526101694915255e-05, | |
| "loss": 1.4918, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 1.9484646320343018, | |
| "learning_rate": 9.492203389830509e-05, | |
| "loss": 1.459, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.489809274673462, | |
| "eval_runtime": 93.0028, | |
| "eval_samples_per_second": 10.752, | |
| "eval_steps_per_second": 2.688, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 1.8115673065185547, | |
| "learning_rate": 9.458305084745763e-05, | |
| "loss": 1.4285, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.803442358970642, | |
| "learning_rate": 9.424406779661016e-05, | |
| "loss": 1.5765, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.2882965803146362, | |
| "learning_rate": 9.39084745762712e-05, | |
| "loss": 1.4512, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.9917691946029663, | |
| "learning_rate": 9.356949152542374e-05, | |
| "loss": 1.5322, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.6726502180099487, | |
| "learning_rate": 9.323050847457627e-05, | |
| "loss": 1.5219, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 3.595461845397949, | |
| "learning_rate": 9.289152542372882e-05, | |
| "loss": 1.444, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 2.5381546020507812, | |
| "learning_rate": 9.255254237288135e-05, | |
| "loss": 1.4474, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 3.1601195335388184, | |
| "learning_rate": 9.22135593220339e-05, | |
| "loss": 1.4111, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 2.147916078567505, | |
| "learning_rate": 9.187457627118645e-05, | |
| "loss": 1.4551, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6436530351638794, | |
| "learning_rate": 9.153559322033899e-05, | |
| "loss": 1.4915, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.458221197128296, | |
| "eval_runtime": 93.0923, | |
| "eval_samples_per_second": 10.742, | |
| "eval_steps_per_second": 2.686, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.2148969173431396, | |
| "learning_rate": 9.119661016949152e-05, | |
| "loss": 1.4816, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 2.3700318336486816, | |
| "learning_rate": 9.085762711864407e-05, | |
| "loss": 1.4501, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 1.5613566637039185, | |
| "learning_rate": 9.05186440677966e-05, | |
| "loss": 1.4646, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 2.175220251083374, | |
| "learning_rate": 9.017966101694916e-05, | |
| "loss": 1.5036, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 12.032076835632324, | |
| "learning_rate": 8.98406779661017e-05, | |
| "loss": 1.4563, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.2785927057266235, | |
| "learning_rate": 8.950169491525424e-05, | |
| "loss": 1.5178, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.5785298347473145, | |
| "learning_rate": 8.916271186440679e-05, | |
| "loss": 1.4257, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.1845338344573975, | |
| "learning_rate": 8.882372881355932e-05, | |
| "loss": 1.4742, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 2.2537214756011963, | |
| "learning_rate": 8.848474576271187e-05, | |
| "loss": 1.4538, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 2.092677354812622, | |
| "learning_rate": 8.814576271186441e-05, | |
| "loss": 1.4436, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.4766532182693481, | |
| "eval_runtime": 93.0059, | |
| "eval_samples_per_second": 10.752, | |
| "eval_steps_per_second": 2.688, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.3880198001861572, | |
| "learning_rate": 8.780677966101696e-05, | |
| "loss": 1.4706, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 1.7497975826263428, | |
| "learning_rate": 8.746779661016949e-05, | |
| "loss": 1.4361, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.1048712730407715, | |
| "learning_rate": 8.713220338983051e-05, | |
| "loss": 1.4484, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 2.206047534942627, | |
| "learning_rate": 8.679322033898306e-05, | |
| "loss": 1.4685, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 2.012110471725464, | |
| "learning_rate": 8.64542372881356e-05, | |
| "loss": 1.429, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 2.0992116928100586, | |
| "learning_rate": 8.611525423728813e-05, | |
| "loss": 1.5028, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 4.423650741577148, | |
| "learning_rate": 8.577627118644068e-05, | |
| "loss": 1.4644, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.7697961330413818, | |
| "learning_rate": 8.543728813559323e-05, | |
| "loss": 1.4537, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 3.925752878189087, | |
| "learning_rate": 8.509830508474576e-05, | |
| "loss": 1.4298, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.335164189338684, | |
| "learning_rate": 8.475932203389832e-05, | |
| "loss": 1.4534, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 1.4249181747436523, | |
| "eval_runtime": 93.056, | |
| "eval_samples_per_second": 10.746, | |
| "eval_steps_per_second": 2.687, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 2.5609335899353027, | |
| "learning_rate": 8.442033898305085e-05, | |
| "loss": 1.4444, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 1.4093282222747803, | |
| "learning_rate": 8.40813559322034e-05, | |
| "loss": 1.4737, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.9211537837982178, | |
| "learning_rate": 8.374237288135593e-05, | |
| "loss": 1.4108, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.1686183214187622, | |
| "learning_rate": 8.340338983050848e-05, | |
| "loss": 1.4603, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 2.1233601570129395, | |
| "learning_rate": 8.306440677966102e-05, | |
| "loss": 1.414, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.9452950954437256, | |
| "learning_rate": 8.272881355932204e-05, | |
| "loss": 1.3891, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 2.8304834365844727, | |
| "learning_rate": 8.238983050847457e-05, | |
| "loss": 1.4634, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.8340227603912354, | |
| "learning_rate": 8.205084745762712e-05, | |
| "loss": 1.4711, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.5472098588943481, | |
| "learning_rate": 8.171186440677967e-05, | |
| "loss": 1.4079, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.8752516508102417, | |
| "learning_rate": 8.137288135593221e-05, | |
| "loss": 1.4255, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.433774471282959, | |
| "eval_runtime": 93.0758, | |
| "eval_samples_per_second": 10.744, | |
| "eval_steps_per_second": 2.686, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 4.256100177764893, | |
| "learning_rate": 8.103389830508476e-05, | |
| "loss": 1.4102, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.966732144355774, | |
| "learning_rate": 8.069491525423729e-05, | |
| "loss": 1.4337, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.9053303599357605, | |
| "learning_rate": 8.035593220338984e-05, | |
| "loss": 1.4517, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 2.6552069187164307, | |
| "learning_rate": 8.001694915254237e-05, | |
| "loss": 1.459, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.2912251949310303, | |
| "learning_rate": 7.967796610169492e-05, | |
| "loss": 1.3857, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 2.4170544147491455, | |
| "learning_rate": 7.933898305084746e-05, | |
| "loss": 1.4032, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.7848442792892456, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.4374, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 4.551065444946289, | |
| "learning_rate": 7.866101694915254e-05, | |
| "loss": 1.4481, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 1.3571339845657349, | |
| "learning_rate": 7.832203389830509e-05, | |
| "loss": 1.4213, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.5277926921844482, | |
| "learning_rate": 7.798305084745763e-05, | |
| "loss": 1.429, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.394229769706726, | |
| "eval_runtime": 93.0895, | |
| "eval_samples_per_second": 10.742, | |
| "eval_steps_per_second": 2.686, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.962740421295166, | |
| "learning_rate": 7.764406779661018e-05, | |
| "loss": 1.4396, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.3276052474975586, | |
| "learning_rate": 7.730508474576271e-05, | |
| "loss": 1.4628, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.9193934202194214, | |
| "learning_rate": 7.696610169491526e-05, | |
| "loss": 1.4234, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.859925627708435, | |
| "learning_rate": 7.662711864406779e-05, | |
| "loss": 1.4427, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 2.7623603343963623, | |
| "learning_rate": 7.628813559322034e-05, | |
| "loss": 1.4046, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 2.3409364223480225, | |
| "learning_rate": 7.594915254237289e-05, | |
| "loss": 1.4351, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.8644472360610962, | |
| "learning_rate": 7.561016949152543e-05, | |
| "loss": 1.4281, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 2.663987398147583, | |
| "learning_rate": 7.527118644067796e-05, | |
| "loss": 1.3743, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.3779089450836182, | |
| "learning_rate": 7.493220338983051e-05, | |
| "loss": 1.4531, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 1.2310881614685059, | |
| "learning_rate": 7.459322033898304e-05, | |
| "loss": 1.427, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.4101628065109253, | |
| "eval_runtime": 92.9756, | |
| "eval_samples_per_second": 10.756, | |
| "eval_steps_per_second": 2.689, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 2.426481246948242, | |
| "learning_rate": 7.425762711864407e-05, | |
| "loss": 1.4215, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 3.706364154815674, | |
| "learning_rate": 7.391864406779662e-05, | |
| "loss": 1.3344, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 2.511159658432007, | |
| "learning_rate": 7.357966101694915e-05, | |
| "loss": 1.4132, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.31131911277771, | |
| "learning_rate": 7.32406779661017e-05, | |
| "loss": 1.4357, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 2.3403029441833496, | |
| "learning_rate": 7.290169491525423e-05, | |
| "loss": 1.455, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.386819362640381, | |
| "learning_rate": 7.256271186440679e-05, | |
| "loss": 1.374, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 12.603606224060059, | |
| "learning_rate": 7.222372881355932e-05, | |
| "loss": 1.461, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 2.5920355319976807, | |
| "learning_rate": 7.188474576271187e-05, | |
| "loss": 1.423, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 3.7238869667053223, | |
| "learning_rate": 7.15457627118644e-05, | |
| "loss": 1.3845, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 4.266007900238037, | |
| "learning_rate": 7.120677966101695e-05, | |
| "loss": 1.4293, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 1.4198664426803589, | |
| "eval_runtime": 93.0273, | |
| "eval_samples_per_second": 10.75, | |
| "eval_steps_per_second": 2.687, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.2583582401275635, | |
| "learning_rate": 7.08677966101695e-05, | |
| "loss": 1.4032, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 3.374390125274658, | |
| "learning_rate": 7.052881355932204e-05, | |
| "loss": 1.3469, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 2.431976318359375, | |
| "learning_rate": 7.018983050847457e-05, | |
| "loss": 1.3861, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 2.953474760055542, | |
| "learning_rate": 6.985084745762712e-05, | |
| "loss": 1.3406, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 1.3457430601119995, | |
| "learning_rate": 6.951186440677967e-05, | |
| "loss": 1.4132, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.713928699493408, | |
| "learning_rate": 6.91728813559322e-05, | |
| "loss": 1.4555, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 3.0444705486297607, | |
| "learning_rate": 6.883389830508476e-05, | |
| "loss": 1.3372, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 2.5544724464416504, | |
| "learning_rate": 6.849491525423729e-05, | |
| "loss": 1.4265, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 3.4634218215942383, | |
| "learning_rate": 6.815593220338984e-05, | |
| "loss": 1.4021, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 2.940263509750366, | |
| "learning_rate": 6.781694915254237e-05, | |
| "loss": 1.3975, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_loss": 1.3409844636917114, | |
| "eval_runtime": 92.9614, | |
| "eval_samples_per_second": 10.757, | |
| "eval_steps_per_second": 2.689, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 2.2668895721435547, | |
| "learning_rate": 6.748135593220339e-05, | |
| "loss": 1.3879, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 2.5967912673950195, | |
| "learning_rate": 6.714237288135593e-05, | |
| "loss": 1.4197, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 4.100217819213867, | |
| "learning_rate": 6.680338983050848e-05, | |
| "loss": 1.4081, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.7082469463348389, | |
| "learning_rate": 6.646440677966101e-05, | |
| "loss": 1.4103, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 2.1128108501434326, | |
| "learning_rate": 6.612542372881356e-05, | |
| "loss": 1.44, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 2.1471731662750244, | |
| "learning_rate": 6.57864406779661e-05, | |
| "loss": 1.412, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 3.2873263359069824, | |
| "learning_rate": 6.544745762711865e-05, | |
| "loss": 1.4407, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.7035319805145264, | |
| "learning_rate": 6.51084745762712e-05, | |
| "loss": 1.3792, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 2.7462477684020996, | |
| "learning_rate": 6.476949152542373e-05, | |
| "loss": 1.3676, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.895967721939087, | |
| "learning_rate": 6.443050847457628e-05, | |
| "loss": 1.4092, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 1.4003005027770996, | |
| "eval_runtime": 93.0106, | |
| "eval_samples_per_second": 10.751, | |
| "eval_steps_per_second": 2.688, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 2.039684295654297, | |
| "learning_rate": 6.409152542372881e-05, | |
| "loss": 1.4015, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.625591516494751, | |
| "learning_rate": 6.375254237288136e-05, | |
| "loss": 1.4362, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.4086389541625977, | |
| "learning_rate": 6.34135593220339e-05, | |
| "loss": 1.4192, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.9734954833984375, | |
| "learning_rate": 6.307457627118645e-05, | |
| "loss": 1.4413, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.1704461574554443, | |
| "learning_rate": 6.273559322033898e-05, | |
| "loss": 1.3856, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.659949541091919, | |
| "learning_rate": 6.239661016949153e-05, | |
| "loss": 1.4044, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.8390787839889526, | |
| "learning_rate": 6.205762711864406e-05, | |
| "loss": 1.3704, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 1.7644896507263184, | |
| "learning_rate": 6.171864406779662e-05, | |
| "loss": 1.4389, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 2.0395686626434326, | |
| "learning_rate": 6.137966101694915e-05, | |
| "loss": 1.397, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.8224962949752808, | |
| "learning_rate": 6.10406779661017e-05, | |
| "loss": 1.351, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.4065361022949219, | |
| "eval_runtime": 93.0021, | |
| "eval_samples_per_second": 10.752, | |
| "eval_steps_per_second": 2.688, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 3.049715995788574, | |
| "learning_rate": 6.070169491525424e-05, | |
| "loss": 1.3315, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 2.025336503982544, | |
| "learning_rate": 6.0362711864406786e-05, | |
| "loss": 1.3722, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 2.9472439289093018, | |
| "learning_rate": 6.002372881355932e-05, | |
| "loss": 1.3185, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 3.2052855491638184, | |
| "learning_rate": 5.968474576271187e-05, | |
| "loss": 1.3029, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 3.9783244132995605, | |
| "learning_rate": 5.9345762711864405e-05, | |
| "loss": 1.3617, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 6.090597152709961, | |
| "learning_rate": 5.900677966101695e-05, | |
| "loss": 1.4188, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 2.486492156982422, | |
| "learning_rate": 5.866779661016949e-05, | |
| "loss": 1.3771, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.2206759452819824, | |
| "learning_rate": 5.832881355932204e-05, | |
| "loss": 1.3583, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.092623710632324, | |
| "learning_rate": 5.7989830508474577e-05, | |
| "loss": 1.344, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 1.9661636352539062, | |
| "learning_rate": 5.765084745762712e-05, | |
| "loss": 1.3364, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 1.3436501026153564, | |
| "eval_runtime": 92.9061, | |
| "eval_samples_per_second": 10.764, | |
| "eval_steps_per_second": 2.691, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 2.299858331680298, | |
| "learning_rate": 5.7311864406779656e-05, | |
| "loss": 1.327, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.999542474746704, | |
| "learning_rate": 5.697288135593221e-05, | |
| "loss": 1.3852, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 4.422431468963623, | |
| "learning_rate": 5.6633898305084755e-05, | |
| "loss": 1.3812, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.9813380241394043, | |
| "learning_rate": 5.629491525423729e-05, | |
| "loss": 1.3517, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.8180012702941895, | |
| "learning_rate": 5.595593220338984e-05, | |
| "loss": 1.3715, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 3.0727899074554443, | |
| "learning_rate": 5.5616949152542374e-05, | |
| "loss": 1.3278, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 1.4579344987869263, | |
| "learning_rate": 5.527796610169492e-05, | |
| "loss": 1.3649, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 1.9728525876998901, | |
| "learning_rate": 5.493898305084746e-05, | |
| "loss": 1.2894, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 3.2770729064941406, | |
| "learning_rate": 5.4600000000000006e-05, | |
| "loss": 1.3748, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 4.356542587280273, | |
| "learning_rate": 5.4261016949152545e-05, | |
| "loss": 1.3153, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.3724653720855713, | |
| "eval_runtime": 92.8763, | |
| "eval_samples_per_second": 10.767, | |
| "eval_steps_per_second": 2.692, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.5841625928878784, | |
| "learning_rate": 5.392203389830509e-05, | |
| "loss": 1.3295, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 8.736273765563965, | |
| "learning_rate": 5.35864406779661e-05, | |
| "loss": 1.3146, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 3.8798937797546387, | |
| "learning_rate": 5.324745762711865e-05, | |
| "loss": 1.2944, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.1168253421783447, | |
| "learning_rate": 5.2908474576271194e-05, | |
| "loss": 1.2995, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 1.7689605951309204, | |
| "learning_rate": 5.256949152542373e-05, | |
| "loss": 1.3166, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 2.00744891166687, | |
| "learning_rate": 5.223050847457628e-05, | |
| "loss": 1.3385, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 1.9501924514770508, | |
| "learning_rate": 5.189152542372881e-05, | |
| "loss": 1.3564, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 2.614849328994751, | |
| "learning_rate": 5.1552542372881366e-05, | |
| "loss": 1.3326, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 2.0598857402801514, | |
| "learning_rate": 5.12135593220339e-05, | |
| "loss": 1.2322, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.12611985206604, | |
| "learning_rate": 5.0874576271186445e-05, | |
| "loss": 1.354, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.368465542793274, | |
| "eval_runtime": 93.0272, | |
| "eval_samples_per_second": 10.75, | |
| "eval_steps_per_second": 2.687, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 1.287361979484558, | |
| "learning_rate": 5.0535593220338984e-05, | |
| "loss": 1.3178, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 3.340104579925537, | |
| "learning_rate": 5.019661016949153e-05, | |
| "loss": 1.3893, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.4543209075927734, | |
| "learning_rate": 4.985762711864407e-05, | |
| "loss": 1.3597, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.6554243564605713, | |
| "learning_rate": 4.951864406779661e-05, | |
| "loss": 1.2903, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.685389518737793, | |
| "learning_rate": 4.917966101694915e-05, | |
| "loss": 1.2902, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 1.9464205503463745, | |
| "learning_rate": 4.88406779661017e-05, | |
| "loss": 1.3333, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 2.7877588272094727, | |
| "learning_rate": 4.850169491525424e-05, | |
| "loss": 1.3016, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.88291335105896, | |
| "learning_rate": 4.816271186440678e-05, | |
| "loss": 1.382, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 4.62669038772583, | |
| "learning_rate": 4.782372881355933e-05, | |
| "loss": 1.3343, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 2.525221824645996, | |
| "learning_rate": 4.748474576271187e-05, | |
| "loss": 1.2822, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 1.3313556909561157, | |
| "eval_runtime": 93.0605, | |
| "eval_samples_per_second": 10.746, | |
| "eval_steps_per_second": 2.686, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 2.8074324131011963, | |
| "learning_rate": 4.714576271186441e-05, | |
| "loss": 1.3487, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 3.902251958847046, | |
| "learning_rate": 4.680677966101695e-05, | |
| "loss": 1.2796, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 3.72412109375, | |
| "learning_rate": 4.646779661016949e-05, | |
| "loss": 1.3161, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 1.7499101161956787, | |
| "learning_rate": 4.612881355932204e-05, | |
| "loss": 1.3477, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 2.252373695373535, | |
| "learning_rate": 4.578983050847458e-05, | |
| "loss": 1.3288, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 3.353426933288574, | |
| "learning_rate": 4.5450847457627125e-05, | |
| "loss": 1.3472, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 2.308408260345459, | |
| "learning_rate": 4.5111864406779664e-05, | |
| "loss": 1.3213, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.369659423828125, | |
| "learning_rate": 4.4772881355932204e-05, | |
| "loss": 1.2767, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 1.9831510782241821, | |
| "learning_rate": 4.443389830508475e-05, | |
| "loss": 1.2989, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 1.2581738233566284, | |
| "learning_rate": 4.409491525423729e-05, | |
| "loss": 1.3224, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 1.3554507493972778, | |
| "eval_runtime": 92.9689, | |
| "eval_samples_per_second": 10.756, | |
| "eval_steps_per_second": 2.689, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 5.903542518615723, | |
| "learning_rate": 4.375593220338983e-05, | |
| "loss": 1.3135, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 2.0982234477996826, | |
| "learning_rate": 4.3416949152542375e-05, | |
| "loss": 1.3036, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.5168031454086304, | |
| "learning_rate": 4.3077966101694915e-05, | |
| "loss": 1.3223, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 2.655869722366333, | |
| "learning_rate": 4.273898305084746e-05, | |
| "loss": 1.3036, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 3.1604421138763428, | |
| "learning_rate": 4.24e-05, | |
| "loss": 1.3172, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 1.9133104085922241, | |
| "learning_rate": 4.206101694915254e-05, | |
| "loss": 1.3596, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 5.270731449127197, | |
| "learning_rate": 4.1722033898305086e-05, | |
| "loss": 1.3021, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 1.63645339012146, | |
| "learning_rate": 4.1383050847457626e-05, | |
| "loss": 1.3197, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 3.240687370300293, | |
| "learning_rate": 4.104406779661017e-05, | |
| "loss": 1.347, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.892014265060425, | |
| "learning_rate": 4.070508474576271e-05, | |
| "loss": 1.2844, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 1.3619292974472046, | |
| "eval_runtime": 92.9703, | |
| "eval_samples_per_second": 10.756, | |
| "eval_steps_per_second": 2.689, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 1.62366783618927, | |
| "learning_rate": 4.036949152542373e-05, | |
| "loss": 1.3391, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 3.9142508506774902, | |
| "learning_rate": 4.0030508474576275e-05, | |
| "loss": 1.2751, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 3.78511381149292, | |
| "learning_rate": 3.9691525423728814e-05, | |
| "loss": 1.3169, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 2.912883758544922, | |
| "learning_rate": 3.9352542372881354e-05, | |
| "loss": 1.3224, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 2.096040725708008, | |
| "learning_rate": 3.90135593220339e-05, | |
| "loss": 1.2708, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 5.406456470489502, | |
| "learning_rate": 3.867457627118644e-05, | |
| "loss": 1.3068, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 1.6677026748657227, | |
| "learning_rate": 3.8335593220338986e-05, | |
| "loss": 1.3202, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 2.0527544021606445, | |
| "learning_rate": 3.7996610169491525e-05, | |
| "loss": 1.2886, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 3.8199574947357178, | |
| "learning_rate": 3.7657627118644065e-05, | |
| "loss": 1.3367, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 2.5120060443878174, | |
| "learning_rate": 3.731864406779661e-05, | |
| "loss": 1.2945, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_loss": 1.3452004194259644, | |
| "eval_runtime": 93.0461, | |
| "eval_samples_per_second": 10.747, | |
| "eval_steps_per_second": 2.687, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 1.0111308097839355, | |
| "learning_rate": 3.697966101694915e-05, | |
| "loss": 1.3713, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.966914176940918, | |
| "learning_rate": 3.66406779661017e-05, | |
| "loss": 1.3124, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 4.211293697357178, | |
| "learning_rate": 3.630508474576271e-05, | |
| "loss": 1.3617, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.8540898561477661, | |
| "learning_rate": 3.596610169491525e-05, | |
| "loss": 1.3393, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 4.376151084899902, | |
| "learning_rate": 3.56271186440678e-05, | |
| "loss": 1.304, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 3.297484874725342, | |
| "learning_rate": 3.528813559322034e-05, | |
| "loss": 1.276, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 3.258236885070801, | |
| "learning_rate": 3.4949152542372885e-05, | |
| "loss": 1.3191, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 4.905130863189697, | |
| "learning_rate": 3.4610169491525425e-05, | |
| "loss": 1.3228, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 2.611107110977173, | |
| "learning_rate": 3.4271186440677964e-05, | |
| "loss": 1.2464, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 4.044171333312988, | |
| "learning_rate": 3.393220338983051e-05, | |
| "loss": 1.2852, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 1.322507619857788, | |
| "eval_runtime": 93.0207, | |
| "eval_samples_per_second": 10.75, | |
| "eval_steps_per_second": 2.688, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 2.803262710571289, | |
| "learning_rate": 3.359322033898305e-05, | |
| "loss": 1.32, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 1.6002579927444458, | |
| "learning_rate": 3.325423728813559e-05, | |
| "loss": 1.2831, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 5.442947864532471, | |
| "learning_rate": 3.2915254237288136e-05, | |
| "loss": 1.2849, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 3.7041471004486084, | |
| "learning_rate": 3.257627118644068e-05, | |
| "loss": 1.2846, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 4.188296794891357, | |
| "learning_rate": 3.223728813559322e-05, | |
| "loss": 1.3026, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 3.6854124069213867, | |
| "learning_rate": 3.189830508474577e-05, | |
| "loss": 1.2429, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 3.1174068450927734, | |
| "learning_rate": 3.155932203389831e-05, | |
| "loss": 1.2728, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 1.2928777933120728, | |
| "learning_rate": 3.1220338983050854e-05, | |
| "loss": 1.291, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 3.058570146560669, | |
| "learning_rate": 3.088135593220339e-05, | |
| "loss": 1.2786, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 3.311408042907715, | |
| "learning_rate": 3.054237288135593e-05, | |
| "loss": 1.2607, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.2871655225753784, | |
| "eval_runtime": 93.0004, | |
| "eval_samples_per_second": 10.753, | |
| "eval_steps_per_second": 2.688, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 3.0149199962615967, | |
| "learning_rate": 3.0203389830508476e-05, | |
| "loss": 1.2897, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 3.8012924194335938, | |
| "learning_rate": 2.986440677966102e-05, | |
| "loss": 1.3003, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.5058399438858032, | |
| "learning_rate": 2.952542372881356e-05, | |
| "loss": 1.245, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 5.085143089294434, | |
| "learning_rate": 2.9186440677966104e-05, | |
| "loss": 1.2702, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.6351230144500732, | |
| "learning_rate": 2.8847457627118647e-05, | |
| "loss": 1.28, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 3.3916802406311035, | |
| "learning_rate": 2.8508474576271187e-05, | |
| "loss": 1.2621, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 2.4449045658111572, | |
| "learning_rate": 2.816949152542373e-05, | |
| "loss": 1.2968, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 2.8779613971710205, | |
| "learning_rate": 2.7830508474576273e-05, | |
| "loss": 1.3546, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 1.7994945049285889, | |
| "learning_rate": 2.7491525423728816e-05, | |
| "loss": 1.2766, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 3.81644344329834, | |
| "learning_rate": 2.7152542372881355e-05, | |
| "loss": 1.2648, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 1.322618007659912, | |
| "eval_runtime": 92.9974, | |
| "eval_samples_per_second": 10.753, | |
| "eval_steps_per_second": 2.688, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 3.8521928787231445, | |
| "learning_rate": 2.6813559322033898e-05, | |
| "loss": 1.2689, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 3.50951886177063, | |
| "learning_rate": 2.647457627118644e-05, | |
| "loss": 1.3133, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 2.187098979949951, | |
| "learning_rate": 2.6135593220338984e-05, | |
| "loss": 1.2842, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 2.276773691177368, | |
| "learning_rate": 2.5796610169491527e-05, | |
| "loss": 1.252, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.8711966276168823, | |
| "learning_rate": 2.5457627118644066e-05, | |
| "loss": 1.2994, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 2.4090774059295654, | |
| "learning_rate": 2.511864406779661e-05, | |
| "loss": 1.2686, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 4.18037748336792, | |
| "learning_rate": 2.4779661016949156e-05, | |
| "loss": 1.318, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.396822452545166, | |
| "learning_rate": 2.4440677966101695e-05, | |
| "loss": 1.27, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 3.4309394359588623, | |
| "learning_rate": 2.4101694915254238e-05, | |
| "loss": 1.2888, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.140385627746582, | |
| "learning_rate": 2.376271186440678e-05, | |
| "loss": 1.2936, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 1.3380318880081177, | |
| "eval_runtime": 93.0298, | |
| "eval_samples_per_second": 10.749, | |
| "eval_steps_per_second": 2.687, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 1.4485564231872559, | |
| "learning_rate": 2.3423728813559324e-05, | |
| "loss": 1.2744, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 2.8173444271087646, | |
| "learning_rate": 2.3084745762711867e-05, | |
| "loss": 1.2774, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 2.9735991954803467, | |
| "learning_rate": 2.2745762711864406e-05, | |
| "loss": 1.3241, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 2.011720657348633, | |
| "learning_rate": 2.240677966101695e-05, | |
| "loss": 1.2912, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 2.2044718265533447, | |
| "learning_rate": 2.2067796610169492e-05, | |
| "loss": 1.2904, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 3.7060163021087646, | |
| "learning_rate": 2.1728813559322035e-05, | |
| "loss": 1.2289, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 4.124302864074707, | |
| "learning_rate": 2.1389830508474575e-05, | |
| "loss": 1.2544, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 3.379523992538452, | |
| "learning_rate": 2.105084745762712e-05, | |
| "loss": 1.2604, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 7.474296569824219, | |
| "learning_rate": 2.0711864406779664e-05, | |
| "loss": 1.274, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.122653007507324, | |
| "learning_rate": 2.0372881355932207e-05, | |
| "loss": 1.2985, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.274635910987854, | |
| "eval_runtime": 93.0906, | |
| "eval_samples_per_second": 10.742, | |
| "eval_steps_per_second": 2.686, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 2.563617467880249, | |
| "learning_rate": 2.0033898305084746e-05, | |
| "loss": 1.1996, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.481621265411377, | |
| "learning_rate": 1.969491525423729e-05, | |
| "loss": 1.2156, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 2.6262869834899902, | |
| "learning_rate": 1.9355932203389832e-05, | |
| "loss": 1.221, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 3.437523603439331, | |
| "learning_rate": 1.9016949152542375e-05, | |
| "loss": 1.1814, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 4.948596477508545, | |
| "learning_rate": 1.8677966101694915e-05, | |
| "loss": 1.1972, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 3.0667967796325684, | |
| "learning_rate": 1.8338983050847458e-05, | |
| "loss": 1.2229, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 2.1678037643432617, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.182, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 2.7943167686462402, | |
| "learning_rate": 1.7661016949152543e-05, | |
| "loss": 1.1889, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.5044032335281372, | |
| "learning_rate": 1.7322033898305083e-05, | |
| "loss": 1.1796, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 3.931037425994873, | |
| "learning_rate": 1.6983050847457626e-05, | |
| "loss": 1.2159, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 1.33604097366333, | |
| "eval_runtime": 93.1418, | |
| "eval_samples_per_second": 10.736, | |
| "eval_steps_per_second": 2.684, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 5.764918804168701, | |
| "learning_rate": 1.6644067796610172e-05, | |
| "loss": 1.1995, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 2.17740535736084, | |
| "learning_rate": 1.6305084745762715e-05, | |
| "loss": 1.1873, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 3.145355224609375, | |
| "learning_rate": 1.5966101694915255e-05, | |
| "loss": 1.2101, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 5.353292942047119, | |
| "learning_rate": 1.563050847457627e-05, | |
| "loss": 1.2076, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 2.399566411972046, | |
| "learning_rate": 1.5291525423728814e-05, | |
| "loss": 1.2099, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 3.8052713871002197, | |
| "learning_rate": 1.4952542372881357e-05, | |
| "loss": 1.1932, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 4.528957843780518, | |
| "learning_rate": 1.4613559322033898e-05, | |
| "loss": 1.1838, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 3.597487449645996, | |
| "learning_rate": 1.4274576271186441e-05, | |
| "loss": 1.187, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 3.2705302238464355, | |
| "learning_rate": 1.3935593220338982e-05, | |
| "loss": 1.1829, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 3.251126289367676, | |
| "learning_rate": 1.3596610169491525e-05, | |
| "loss": 1.1888, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 1.3534026145935059, | |
| "eval_runtime": 93.1511, | |
| "eval_samples_per_second": 10.735, | |
| "eval_steps_per_second": 2.684, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 3.9193084239959717, | |
| "learning_rate": 1.325762711864407e-05, | |
| "loss": 1.2176, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 1.7747070789337158, | |
| "learning_rate": 1.2918644067796613e-05, | |
| "loss": 1.1845, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.7094638347625732, | |
| "learning_rate": 1.2579661016949154e-05, | |
| "loss": 1.2025, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 4.238964557647705, | |
| "learning_rate": 1.2240677966101695e-05, | |
| "loss": 1.1665, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 1.669313669204712, | |
| "learning_rate": 1.1901694915254238e-05, | |
| "loss": 1.1792, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 8.226383209228516, | |
| "learning_rate": 1.1562711864406781e-05, | |
| "loss": 1.1591, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 3.0040552616119385, | |
| "learning_rate": 1.1223728813559322e-05, | |
| "loss": 1.2112, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 1.9113960266113281, | |
| "learning_rate": 1.0884745762711865e-05, | |
| "loss": 1.164, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 2.2811169624328613, | |
| "learning_rate": 1.0545762711864406e-05, | |
| "loss": 1.1807, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 3.2554876804351807, | |
| "learning_rate": 1.020677966101695e-05, | |
| "loss": 1.2362, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 1.315274715423584, | |
| "eval_runtime": 93.1384, | |
| "eval_samples_per_second": 10.737, | |
| "eval_steps_per_second": 2.684, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 3.656125545501709, | |
| "learning_rate": 9.867796610169492e-06, | |
| "loss": 1.1712, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 4.444190502166748, | |
| "learning_rate": 9.528813559322035e-06, | |
| "loss": 1.2532, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 5.304549217224121, | |
| "learning_rate": 9.189830508474576e-06, | |
| "loss": 1.1848, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 4.003425121307373, | |
| "learning_rate": 8.85084745762712e-06, | |
| "loss": 1.1601, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 4.701766490936279, | |
| "learning_rate": 8.511864406779662e-06, | |
| "loss": 1.1682, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 4.16826057434082, | |
| "learning_rate": 8.172881355932203e-06, | |
| "loss": 1.1806, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 3.569227695465088, | |
| "learning_rate": 7.833898305084746e-06, | |
| "loss": 1.1295, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 6.8756890296936035, | |
| "learning_rate": 7.494915254237289e-06, | |
| "loss": 1.1863, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 4.841982364654541, | |
| "learning_rate": 7.155932203389831e-06, | |
| "loss": 1.1651, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 6.766154766082764, | |
| "learning_rate": 6.816949152542373e-06, | |
| "loss": 1.1831, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.3225314617156982, | |
| "eval_runtime": 93.1704, | |
| "eval_samples_per_second": 10.733, | |
| "eval_steps_per_second": 2.683, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 4.122812747955322, | |
| "learning_rate": 6.477966101694915e-06, | |
| "loss": 1.1934, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 2.2169628143310547, | |
| "learning_rate": 6.1389830508474575e-06, | |
| "loss": 1.1044, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 3.0423035621643066, | |
| "learning_rate": 5.8e-06, | |
| "loss": 1.1876, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 3.517133951187134, | |
| "learning_rate": 5.4610169491525424e-06, | |
| "loss": 1.2411, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 3.3230624198913574, | |
| "learning_rate": 5.1220338983050845e-06, | |
| "loss": 1.189, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 5.101288795471191, | |
| "learning_rate": 4.7830508474576274e-06, | |
| "loss": 1.1435, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 3.061377763748169, | |
| "learning_rate": 4.4440677966101695e-06, | |
| "loss": 1.1902, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 3.9341752529144287, | |
| "learning_rate": 4.1050847457627124e-06, | |
| "loss": 1.22, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 3.691490888595581, | |
| "learning_rate": 3.766101694915254e-06, | |
| "loss": 1.2105, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 3.67085599899292, | |
| "learning_rate": 3.427118644067797e-06, | |
| "loss": 1.1909, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 1.294958472251892, | |
| "eval_runtime": 93.09, | |
| "eval_samples_per_second": 10.742, | |
| "eval_steps_per_second": 2.686, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 4.373409748077393, | |
| "learning_rate": 3.088135593220339e-06, | |
| "loss": 1.2423, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 4.1964006423950195, | |
| "learning_rate": 2.7491525423728816e-06, | |
| "loss": 1.1482, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 4.233890056610107, | |
| "learning_rate": 2.4135593220338984e-06, | |
| "loss": 1.1824, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 3.9279072284698486, | |
| "learning_rate": 2.0745762711864404e-06, | |
| "loss": 1.1466, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 3.39454984664917, | |
| "learning_rate": 1.7355932203389831e-06, | |
| "loss": 1.1302, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 1.3635748624801636, | |
| "learning_rate": 1.3966101694915254e-06, | |
| "loss": 1.214, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 3.5587875843048096, | |
| "learning_rate": 1.057627118644068e-06, | |
| "loss": 1.1981, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 7.449352741241455, | |
| "learning_rate": 7.186440677966102e-07, | |
| "loss": 1.1196, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 2.7902631759643555, | |
| "learning_rate": 3.796610169491526e-07, | |
| "loss": 1.1452, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.423699855804443, | |
| "learning_rate": 4.067796610169492e-08, | |
| "loss": 1.197, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 1.2965623140335083, | |
| "eval_runtime": 93.1115, | |
| "eval_samples_per_second": 10.74, | |
| "eval_steps_per_second": 2.685, | |
| "step": 30000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 2000, | |
| "total_flos": 4.8306377981952e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |