| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.695652173913043, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06956521739130435, | |
| "grad_norm": 0.027621353045105934, | |
| "learning_rate": 1.9989930665413148e-05, | |
| "loss": 0.6357, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1391304347826087, | |
| "grad_norm": 0.027610378339886665, | |
| "learning_rate": 1.9959742939952393e-05, | |
| "loss": 0.6181, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 0.029305126518011093, | |
| "learning_rate": 1.990949761767935e-05, | |
| "loss": 0.642, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 0.028681788593530655, | |
| "learning_rate": 1.98392958859863e-05, | |
| "loss": 0.646, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.028845500200986862, | |
| "learning_rate": 1.9749279121818235e-05, | |
| "loss": 0.6416, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 0.03184295445680618, | |
| "learning_rate": 1.9639628606958535e-05, | |
| "loss": 0.6625, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.48695652173913045, | |
| "grad_norm": 0.030502479523420334, | |
| "learning_rate": 1.9510565162951538e-05, | |
| "loss": 0.645, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 0.030076082795858383, | |
| "learning_rate": 1.9362348706397374e-05, | |
| "loss": 0.6285, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "grad_norm": 0.032947439700365067, | |
| "learning_rate": 1.919527772551451e-05, | |
| "loss": 0.6641, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.03192123770713806, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.6508, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7652173913043478, | |
| "grad_norm": 0.0321270152926445, | |
| "learning_rate": 1.880595531856738e-05, | |
| "loss": 0.6436, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 0.03189520537853241, | |
| "learning_rate": 1.8584487936018663e-05, | |
| "loss": 0.6653, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9043478260869565, | |
| "grad_norm": 0.03357314690947533, | |
| "learning_rate": 1.834573253721303e-05, | |
| "loss": 0.636, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9739130434782609, | |
| "grad_norm": 0.031242311000823975, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 0.6537, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.03731789439916611, | |
| "learning_rate": 1.78183148246803e-05, | |
| "loss": 0.6894, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6298047304153442, | |
| "eval_runtime": 2.6261, | |
| "eval_samples_per_second": 1.523, | |
| "eval_steps_per_second": 1.523, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0695652173913044, | |
| "grad_norm": 0.0358547680079937, | |
| "learning_rate": 1.7530714660036112e-05, | |
| "loss": 0.6516, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1391304347826088, | |
| "grad_norm": 0.0324733592569828, | |
| "learning_rate": 1.7227948638273918e-05, | |
| "loss": 0.6244, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.208695652173913, | |
| "grad_norm": 0.03361370787024498, | |
| "learning_rate": 1.691062648986865e-05, | |
| "loss": 0.6417, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.2782608695652173, | |
| "grad_norm": 0.033928435295820236, | |
| "learning_rate": 1.657938725939713e-05, | |
| "loss": 0.6601, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.3478260869565217, | |
| "grad_norm": 0.03608125448226929, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6532, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.4173913043478261, | |
| "grad_norm": 0.03397082909941673, | |
| "learning_rate": 1.5877852522924733e-05, | |
| "loss": 0.6205, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.4869565217391305, | |
| "grad_norm": 0.03406910225749016, | |
| "learning_rate": 1.5508969814521026e-05, | |
| "loss": 0.6256, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.5565217391304347, | |
| "grad_norm": 0.034135762602090836, | |
| "learning_rate": 1.5128992774059063e-05, | |
| "loss": 0.6332, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.626086956521739, | |
| "grad_norm": 0.03473026305437088, | |
| "learning_rate": 1.4738686624729987e-05, | |
| "loss": 0.6561, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.6956521739130435, | |
| "grad_norm": 0.03459230437874794, | |
| "learning_rate": 1.4338837391175582e-05, | |
| "loss": 0.6391, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.7652173913043478, | |
| "grad_norm": 0.0341939777135849, | |
| "learning_rate": 1.3930250316539237e-05, | |
| "loss": 0.6393, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.8347826086956522, | |
| "grad_norm": 0.03445998951792717, | |
| "learning_rate": 1.3513748240813429e-05, | |
| "loss": 0.6429, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.9043478260869566, | |
| "grad_norm": 0.03501657024025917, | |
| "learning_rate": 1.3090169943749475e-05, | |
| "loss": 0.6483, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.973913043478261, | |
| "grad_norm": 0.03379024565219879, | |
| "learning_rate": 1.2660368455666752e-05, | |
| "loss": 0.6247, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.034777939319610596, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.6324, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.620289146900177, | |
| "eval_runtime": 2.6255, | |
| "eval_samples_per_second": 1.524, | |
| "eval_steps_per_second": 1.524, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0695652173913044, | |
| "grad_norm": 0.035359956324100494, | |
| "learning_rate": 1.1785568947986368e-05, | |
| "loss": 0.6416, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.139130434782609, | |
| "grad_norm": 0.03541162610054016, | |
| "learning_rate": 1.1342332658176556e-05, | |
| "loss": 0.6432, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.208695652173913, | |
| "grad_norm": 0.03285665437579155, | |
| "learning_rate": 1.0896393089034336e-05, | |
| "loss": 0.6128, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.2782608695652176, | |
| "grad_norm": 0.03517802432179451, | |
| "learning_rate": 1.044864830350515e-05, | |
| "loss": 0.6557, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.3478260869565215, | |
| "grad_norm": 0.03373734652996063, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6298, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.417391304347826, | |
| "grad_norm": 0.0340682752430439, | |
| "learning_rate": 9.551351696494854e-06, | |
| "loss": 0.6192, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.4869565217391303, | |
| "grad_norm": 0.03553012013435364, | |
| "learning_rate": 9.103606910965666e-06, | |
| "loss": 0.637, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.5565217391304347, | |
| "grad_norm": 0.036372989416122437, | |
| "learning_rate": 8.657667341823449e-06, | |
| "loss": 0.6286, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.626086956521739, | |
| "grad_norm": 0.03410479426383972, | |
| "learning_rate": 8.214431052013636e-06, | |
| "loss": 0.602, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.6956521739130435, | |
| "grad_norm": 0.035654351115226746, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.6282, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.765217391304348, | |
| "grad_norm": 0.03418293967843056, | |
| "learning_rate": 7.33963154433325e-06, | |
| "loss": 0.6384, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.8347826086956522, | |
| "grad_norm": 0.035010650753974915, | |
| "learning_rate": 6.909830056250527e-06, | |
| "loss": 0.6232, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.9043478260869566, | |
| "grad_norm": 0.03416571021080017, | |
| "learning_rate": 6.486251759186573e-06, | |
| "loss": 0.6533, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.973913043478261, | |
| "grad_norm": 0.03395598754286766, | |
| "learning_rate": 6.069749683460765e-06, | |
| "loss": 0.6205, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.0333721823990345, | |
| "learning_rate": 5.66116260882442e-06, | |
| "loss": 0.6163, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6122885942459106, | |
| "eval_runtime": 2.6295, | |
| "eval_samples_per_second": 1.521, | |
| "eval_steps_per_second": 1.521, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0695652173913044, | |
| "grad_norm": 0.033700425177812576, | |
| "learning_rate": 5.2613133752700145e-06, | |
| "loss": 0.6408, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.139130434782609, | |
| "grad_norm": 0.03324361890554428, | |
| "learning_rate": 4.87100722594094e-06, | |
| "loss": 0.62, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 3.208695652173913, | |
| "grad_norm": 0.03405730798840523, | |
| "learning_rate": 4.491030185478976e-06, | |
| "loss": 0.6257, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.2782608695652176, | |
| "grad_norm": 0.03492354601621628, | |
| "learning_rate": 4.12214747707527e-06, | |
| "loss": 0.6349, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 3.3478260869565215, | |
| "grad_norm": 0.034349583089351654, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.6282, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.417391304347826, | |
| "grad_norm": 0.03548488765954971, | |
| "learning_rate": 3.4206127406028744e-06, | |
| "loss": 0.6427, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.4869565217391303, | |
| "grad_norm": 0.03456812724471092, | |
| "learning_rate": 3.089373510131354e-06, | |
| "loss": 0.62, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.5565217391304347, | |
| "grad_norm": 0.03488166257739067, | |
| "learning_rate": 2.7720513617260857e-06, | |
| "loss": 0.6263, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 3.626086956521739, | |
| "grad_norm": 0.0336606539785862, | |
| "learning_rate": 2.469285339963892e-06, | |
| "loss": 0.63, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.6956521739130435, | |
| "grad_norm": 0.03267661854624748, | |
| "learning_rate": 2.1816851753197023e-06, | |
| "loss": 0.6297, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.765217391304348, | |
| "grad_norm": 0.0337102934718132, | |
| "learning_rate": 1.9098300562505266e-06, | |
| "loss": 0.5945, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.8347826086956522, | |
| "grad_norm": 0.035517044365406036, | |
| "learning_rate": 1.6542674627869738e-06, | |
| "loss": 0.6343, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.9043478260869566, | |
| "grad_norm": 0.03479280695319176, | |
| "learning_rate": 1.4155120639813392e-06, | |
| "loss": 0.6163, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.973913043478261, | |
| "grad_norm": 0.035013146698474884, | |
| "learning_rate": 1.19404468143262e-06, | |
| "loss": 0.6114, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.03396342694759369, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.6191, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.609683632850647, | |
| "eval_runtime": 2.6229, | |
| "eval_samples_per_second": 1.525, | |
| "eval_steps_per_second": 1.525, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.069565217391304, | |
| "grad_norm": 0.03345245122909546, | |
| "learning_rate": 8.047222744854943e-07, | |
| "loss": 0.6416, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 4.139130434782609, | |
| "grad_norm": 0.03468826040625572, | |
| "learning_rate": 6.37651293602628e-07, | |
| "loss": 0.6269, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 4.208695652173913, | |
| "grad_norm": 0.031891003251075745, | |
| "learning_rate": 4.894348370484648e-07, | |
| "loss": 0.6147, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 4.278260869565218, | |
| "grad_norm": 0.033924877643585205, | |
| "learning_rate": 3.603713930414676e-07, | |
| "loss": 0.611, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 0.033348944038152695, | |
| "learning_rate": 2.507208781817638e-07, | |
| "loss": 0.6372, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 4.417391304347826, | |
| "grad_norm": 0.03544296696782112, | |
| "learning_rate": 1.6070411401370335e-07, | |
| "loss": 0.6176, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.48695652173913, | |
| "grad_norm": 0.03346049040555954, | |
| "learning_rate": 9.0502382320653e-08, | |
| "loss": 0.6282, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 4.556521739130435, | |
| "grad_norm": 0.03482738509774208, | |
| "learning_rate": 4.025706004760932e-08, | |
| "loss": 0.6432, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.626086956521739, | |
| "grad_norm": 0.03497692570090294, | |
| "learning_rate": 1.0069334586854106e-08, | |
| "loss": 0.6121, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.695652173913043, | |
| "grad_norm": 0.034334901720285416, | |
| "learning_rate": 0.0, | |
| "loss": 0.6054, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.695652173913043, | |
| "eval_loss": 0.609480619430542, | |
| "eval_runtime": 2.628, | |
| "eval_samples_per_second": 1.522, | |
| "eval_steps_per_second": 1.522, | |
| "step": 70 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 1.657443415228416e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |