| { | |
| "best_metric": 0.1841961294412613, | |
| "best_model_checkpoint": "llama-2-7b-hf-text2image-prompts-Liege/checkpoint-65", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 65, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.0005, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9403, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.9234, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.9738, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.9561, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.9102, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 0.8407344222068787, | |
| "eval_runtime": 5.7929, | |
| "eval_samples_per_second": 0.691, | |
| "eval_steps_per_second": 0.691, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.9346, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.8937, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8569, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00019753086419753085, | |
| "loss": 0.8014, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00019506172839506175, | |
| "loss": 0.7843, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001925925925925926, | |
| "loss": 0.7857, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00019012345679012346, | |
| "loss": 0.711, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6122090816497803, | |
| "eval_runtime": 5.6653, | |
| "eval_samples_per_second": 0.706, | |
| "eval_steps_per_second": 0.706, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00018765432098765433, | |
| "loss": 0.6708, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0001851851851851852, | |
| "loss": 0.6309, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.00018271604938271605, | |
| "loss": 0.5825, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.00018024691358024692, | |
| "loss": 0.548, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.5468, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00017530864197530866, | |
| "loss": 0.4747, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 0.41001003980636597, | |
| "eval_runtime": 5.6712, | |
| "eval_samples_per_second": 0.705, | |
| "eval_steps_per_second": 0.705, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.0001728395061728395, | |
| "loss": 0.4748, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00017037037037037037, | |
| "loss": 0.4252, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.00016790123456790124, | |
| "loss": 0.3978, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0001654320987654321, | |
| "loss": 0.361, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.00016296296296296295, | |
| "loss": 0.3544, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.00016049382716049385, | |
| "loss": 0.3317, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0001580246913580247, | |
| "loss": 0.2919, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.2563404142856598, | |
| "eval_runtime": 5.6469, | |
| "eval_samples_per_second": 0.708, | |
| "eval_steps_per_second": 0.708, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.2837, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.0001530864197530864, | |
| "loss": 0.2923, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 0.0001506172839506173, | |
| "loss": 0.2485, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 0.00014814814814814815, | |
| "loss": 0.2767, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 0.00014567901234567902, | |
| "loss": 0.2626, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 0.00014320987654320989, | |
| "loss": 0.2513, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_loss": 0.22316063940525055, | |
| "eval_runtime": 5.6083, | |
| "eval_samples_per_second": 0.713, | |
| "eval_steps_per_second": 0.713, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.00014074074074074076, | |
| "loss": 0.2462, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 0.0001382716049382716, | |
| "loss": 0.2339, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00013580246913580247, | |
| "loss": 0.2269, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.2558, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.0001308641975308642, | |
| "loss": 0.2005, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 0.00012839506172839505, | |
| "loss": 0.2104, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00012592592592592592, | |
| "loss": 0.2161, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.20953989028930664, | |
| "eval_runtime": 5.6454, | |
| "eval_samples_per_second": 0.709, | |
| "eval_steps_per_second": 0.709, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.0001234567901234568, | |
| "loss": 0.2015, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 0.00012098765432098766, | |
| "loss": 0.2465, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.00011851851851851852, | |
| "loss": 0.194, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.00011604938271604939, | |
| "loss": 0.191, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.00011358024691358025, | |
| "loss": 0.2012, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.1807, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "eval_loss": 0.2051384598016739, | |
| "eval_runtime": 5.64, | |
| "eval_samples_per_second": 0.709, | |
| "eval_steps_per_second": 0.709, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.00010864197530864197, | |
| "loss": 0.2073, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 0.00010617283950617284, | |
| "loss": 0.1691, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 0.0001037037037037037, | |
| "loss": 0.1751, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 0.00010123456790123458, | |
| "loss": 0.1844, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 9.876543209876543e-05, | |
| "loss": 0.1975, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 9.62962962962963e-05, | |
| "loss": 0.1567, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.382716049382717e-05, | |
| "loss": 0.1815, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.2019922435283661, | |
| "eval_runtime": 5.629, | |
| "eval_samples_per_second": 0.711, | |
| "eval_steps_per_second": 0.711, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 9.135802469135802e-05, | |
| "loss": 0.1686, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.1496, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 8.641975308641975e-05, | |
| "loss": 0.1446, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.395061728395062e-05, | |
| "loss": 0.1649, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 8.148148148148148e-05, | |
| "loss": 0.1539, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "learning_rate": 7.901234567901235e-05, | |
| "loss": 0.1652, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "eval_loss": 0.1942683905363083, | |
| "eval_runtime": 5.6323, | |
| "eval_samples_per_second": 0.71, | |
| "eval_steps_per_second": 0.71, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 7.65432098765432e-05, | |
| "loss": 0.1352, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 0.1361, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 7.160493827160494e-05, | |
| "loss": 0.1214, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 6.91358024691358e-05, | |
| "loss": 0.1329, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.1072, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 6.419753086419753e-05, | |
| "loss": 0.1123, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 6.17283950617284e-05, | |
| "loss": 0.13, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.1841961294412613, | |
| "eval_runtime": 5.6483, | |
| "eval_samples_per_second": 0.708, | |
| "eval_steps_per_second": 0.708, | |
| "step": 65 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 90, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 1.056823528390656e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |