| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.620745542949757, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.5733333333333334, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.88, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.6499700198395969, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.7025343915343916, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.5733333333333334, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.6933983700329899, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.5733333333333334, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.21599999999999997, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.44222222222222224, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.8077777777777778, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.66, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.94, | |
| "eval_NanoFEVER_cosine_map@100": 0.7456105053991163, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.7750000000000001, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.66, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.7909186074531693, | |
| "eval_NanoFEVER_cosine_precision@1": 0.66, | |
| "eval_NanoFEVER_cosine_precision@5": 0.19999999999999996, | |
| "eval_NanoFEVER_cosine_recall@1": 0.6166666666666667, | |
| "eval_NanoFEVER_cosine_recall@5": 0.9133333333333333, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.7, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.92, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.659131307951428, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.795, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.7, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.7005437199248264, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.7, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.292, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.35, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.73, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.36, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.78, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.5451682461682461, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.5376031746031745, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.36, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.5887327827209738, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.36, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.156, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.36, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.78, | |
| "eval_mldr_loss": 98.74517822265625, | |
| "eval_mldr_runtime": 49.7129, | |
| "eval_mldr_samples_per_second": 10.058, | |
| "eval_mldr_steps_per_second": 0.644, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0, | |
| "eval_squad_loss": 4.950023174285889, | |
| "eval_squad_runtime": 0.9354, | |
| "eval_squad_samples_per_second": 51.314, | |
| "eval_squad_steps_per_second": 3.207, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0, | |
| "eval_narrative_qa_loss": 49.38987731933594, | |
| "eval_narrative_qa_runtime": 1.0306, | |
| "eval_narrative_qa_samples_per_second": 290.117, | |
| "eval_narrative_qa_steps_per_second": 18.436, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.01620745542949757, | |
| "grad_norm": 1336.0, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 86.6132, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03241491085899514, | |
| "grad_norm": 828.0, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 83.7595, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04862236628849271, | |
| "grad_norm": 388.0, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 65.5413, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06482982171799027, | |
| "grad_norm": 320.0, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 40.6319, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08103727714748785, | |
| "grad_norm": 296.0, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 26.2768, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09724473257698542, | |
| "grad_norm": 290.0, | |
| "learning_rate": 4.9997781212386585e-05, | |
| "loss": 14.4879, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11345218800648298, | |
| "grad_norm": 160.0, | |
| "learning_rate": 4.9980033274458193e-05, | |
| "loss": 9.4888, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12965964343598055, | |
| "grad_norm": 212.0, | |
| "learning_rate": 4.994454999929178e-05, | |
| "loss": 8.831, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1458670988654781, | |
| "grad_norm": 109.0, | |
| "learning_rate": 4.98913565793218e-05, | |
| "loss": 7.6776, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1620745542949757, | |
| "grad_norm": 71.5, | |
| "learning_rate": 4.982049078084071e-05, | |
| "loss": 6.4716, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1620745542949757, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.32666666666666666, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.6333333333333333, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.39934768753390587, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.4495740740740741, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.32666666666666666, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.43241374431385005, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.32666666666666666, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.13466666666666668, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.25666666666666665, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.5466666666666667, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.24, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.66, | |
| "eval_NanoFEVER_cosine_map@100": 0.3935747452453245, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.39322222222222225, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.24, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.4387408055008589, | |
| "eval_NanoFEVER_cosine_precision@1": 0.24, | |
| "eval_NanoFEVER_cosine_precision@5": 0.132, | |
| "eval_NanoFEVER_cosine_recall@1": 0.24, | |
| "eval_NanoFEVER_cosine_recall@5": 0.63, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.42, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.58, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.3274226457297244, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.4911666666666667, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.42, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.35120107956309937, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.42, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.14, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.21, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.35, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.66, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.47704567162666855, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.4643333333333334, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.5072993478775919, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.132, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.66, | |
| "eval_mldr_loss": 4.950920581817627, | |
| "eval_mldr_runtime": 14.3444, | |
| "eval_mldr_samples_per_second": 34.857, | |
| "eval_mldr_steps_per_second": 2.231, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1620745542949757, | |
| "eval_squad_loss": 2.0146381855010986, | |
| "eval_squad_runtime": 0.9385, | |
| "eval_squad_samples_per_second": 51.146, | |
| "eval_squad_steps_per_second": 3.197, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1620745542949757, | |
| "eval_narrative_qa_loss": 4.486134052276611, | |
| "eval_narrative_qa_runtime": 1.1506, | |
| "eval_narrative_qa_samples_per_second": 259.856, | |
| "eval_narrative_qa_steps_per_second": 16.513, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17828200972447325, | |
| "grad_norm": 107.5, | |
| "learning_rate": 4.973200291718561e-05, | |
| "loss": 6.1959, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19448946515397084, | |
| "grad_norm": 50.75, | |
| "learning_rate": 4.962595581301673e-05, | |
| "loss": 4.5491, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2106969205834684, | |
| "grad_norm": 50.25, | |
| "learning_rate": 4.95024247597132e-05, | |
| "loss": 4.0871, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.22690437601296595, | |
| "grad_norm": 53.25, | |
| "learning_rate": 4.936149746191758e-05, | |
| "loss": 3.0861, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24311183144246354, | |
| "grad_norm": 25.5, | |
| "learning_rate": 4.920327397526731e-05, | |
| "loss": 2.7792, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2593192868719611, | |
| "grad_norm": 122.0, | |
| "learning_rate": 4.9027866635357136e-05, | |
| "loss": 2.2107, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2755267423014587, | |
| "grad_norm": 410.0, | |
| "learning_rate": 4.883539997798303e-05, | |
| "loss": 1.6997, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2917341977309562, | |
| "grad_norm": 66.0, | |
| "learning_rate": 4.8626010650724204e-05, | |
| "loss": 1.4911, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3079416531604538, | |
| "grad_norm": 1072.0, | |
| "learning_rate": 4.8399847315926e-05, | |
| "loss": 1.6304, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3241491085899514, | |
| "grad_norm": 16.5, | |
| "learning_rate": 4.815707054515248e-05, | |
| "loss": 1.5042, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3241491085899514, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.24666666666666667, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.4533333333333333, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.2917657022707883, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.3391984126984127, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.24666666666666667, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.3047832475255773, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.24666666666666667, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.09466666666666669, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.18333333333333335, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.37333333333333335, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.08, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.26, | |
| "eval_NanoFEVER_cosine_map@100": 0.15831046306304816, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.1541904761904762, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.08, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.15530091408199076, | |
| "eval_NanoFEVER_cosine_precision@1": 0.08, | |
| "eval_NanoFEVER_cosine_precision@5": 0.052000000000000005, | |
| "eval_NanoFEVER_cosine_recall@1": 0.07, | |
| "eval_NanoFEVER_cosine_recall@5": 0.24, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.36, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.5, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.2657203181494037, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.4257142857142857, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.36, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.2891308855457201, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.36, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.11200000000000003, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.18, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.28, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.6, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.45126632559991314, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.4376904761904762, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.46991794294902106, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.12000000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.6, | |
| "eval_mldr_loss": 1.3893917798995972, | |
| "eval_mldr_runtime": 14.4056, | |
| "eval_mldr_samples_per_second": 34.709, | |
| "eval_mldr_steps_per_second": 2.221, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3241491085899514, | |
| "eval_squad_loss": 0.7646868824958801, | |
| "eval_squad_runtime": 1.072, | |
| "eval_squad_samples_per_second": 44.776, | |
| "eval_squad_steps_per_second": 2.799, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3241491085899514, | |
| "eval_narrative_qa_loss": 2.59745192527771, | |
| "eval_narrative_qa_runtime": 1.1589, | |
| "eval_narrative_qa_samples_per_second": 257.994, | |
| "eval_narrative_qa_steps_per_second": 16.394, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34035656401944897, | |
| "grad_norm": 35.0, | |
| "learning_rate": 4.7897852705183785e-05, | |
| "loss": 1.6009, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3565640194489465, | |
| "grad_norm": 1352.0, | |
| "learning_rate": 4.7622377835639064e-05, | |
| "loss": 1.5156, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3727714748784441, | |
| "grad_norm": 704.0, | |
| "learning_rate": 4.73308415183119e-05, | |
| "loss": 1.5305, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3889789303079417, | |
| "grad_norm": 824.0, | |
| "learning_rate": 4.702345073831109e-05, | |
| "loss": 1.6842, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4051863857374392, | |
| "grad_norm": 21760.0, | |
| "learning_rate": 4.6700423737105236e-05, | |
| "loss": 1.2207, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4213938411669368, | |
| "grad_norm": 3040.0, | |
| "learning_rate": 4.63619898575755e-05, | |
| "loss": 1.2634, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4376012965964344, | |
| "grad_norm": 46.5, | |
| "learning_rate": 4.600838938118672e-05, | |
| "loss": 1.3175, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4538087520259319, | |
| "grad_norm": 6112.0, | |
| "learning_rate": 4.563987335739216e-05, | |
| "loss": 1.3107, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.4700162074554295, | |
| "grad_norm": 2256.0, | |
| "learning_rate": 4.525670342539332e-05, | |
| "loss": 1.3163, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4862236628849271, | |
| "grad_norm": 860.0, | |
| "learning_rate": 4.485915162838122e-05, | |
| "loss": 1.3703, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4862236628849271, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.37333333333333335, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.24055436231400043, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.26788624338624334, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.24938490224868604, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.07733333333333335, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.15, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.31666666666666665, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.06, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.18, | |
| "eval_NanoFEVER_cosine_map@100": 0.11320690658806287, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.11538888888888887, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.06, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.11246425246579403, | |
| "eval_NanoFEVER_cosine_precision@1": 0.06, | |
| "eval_NanoFEVER_cosine_precision@5": 0.036000000000000004, | |
| "eval_NanoFEVER_cosine_recall@1": 0.05, | |
| "eval_NanoFEVER_cosine_recall@5": 0.17, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.36, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.17264044237203358, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.26374603174603173, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.18589124040083965, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.08, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.1, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.2, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.58, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4358157379819049, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.42452380952380947, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.44979921387942445, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.11600000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.58, | |
| "eval_mldr_loss": 1.218342900276184, | |
| "eval_mldr_runtime": 14.4679, | |
| "eval_mldr_samples_per_second": 34.559, | |
| "eval_mldr_steps_per_second": 2.212, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4862236628849271, | |
| "eval_squad_loss": 1.0243560075759888, | |
| "eval_squad_runtime": 0.9647, | |
| "eval_squad_samples_per_second": 49.755, | |
| "eval_squad_steps_per_second": 3.11, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4862236628849271, | |
| "eval_narrative_qa_loss": 2.635084629058838, | |
| "eval_narrative_qa_runtime": 1.1616, | |
| "eval_narrative_qa_samples_per_second": 257.409, | |
| "eval_narrative_qa_steps_per_second": 16.357, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5024311183144247, | |
| "grad_norm": 39.25, | |
| "learning_rate": 4.444750022039099e-05, | |
| "loss": 1.2901, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5186385737439222, | |
| "grad_norm": 260.0, | |
| "learning_rate": 4.4022041465907036e-05, | |
| "loss": 1.245, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5348460291734197, | |
| "grad_norm": 2192.0, | |
| "learning_rate": 4.358307743236092e-05, | |
| "loss": 1.1749, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5510534846029174, | |
| "grad_norm": 1064.0, | |
| "learning_rate": 4.3130919775669374e-05, | |
| "loss": 1.0522, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5672609400324149, | |
| "grad_norm": 1120.0, | |
| "learning_rate": 4.2665889518964684e-05, | |
| "loss": 1.1319, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5834683954619124, | |
| "grad_norm": 1552.0, | |
| "learning_rate": 4.2188316824674504e-05, | |
| "loss": 1.1986, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5996758508914101, | |
| "grad_norm": 133.0, | |
| "learning_rate": 4.169854076011292e-05, | |
| "loss": 1.3508, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6158833063209076, | |
| "grad_norm": 5408.0, | |
| "learning_rate": 4.119690905674937e-05, | |
| "loss": 1.3465, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6320907617504052, | |
| "grad_norm": 1624.0, | |
| "learning_rate": 4.068377786332593e-05, | |
| "loss": 1.3062, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6482982171799028, | |
| "grad_norm": 1192.0, | |
| "learning_rate": 4.0159511492998746e-05, | |
| "loss": 0.9974, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6482982171799028, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.20000000000000004, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.39333333333333337, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.25566257669608067, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.2828835978835979, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.20000000000000004, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.2711361604281084, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.20000000000000004, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.08266666666666668, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.16333333333333333, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.3433333333333333, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.08, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.2, | |
| "eval_NanoFEVER_cosine_map@100": 0.12628338770667222, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.1275, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.08, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.1285946755597121, | |
| "eval_NanoFEVER_cosine_precision@1": 0.08, | |
| "eval_NanoFEVER_cosine_precision@5": 0.04, | |
| "eval_NanoFEVER_cosine_recall@1": 0.07, | |
| "eval_NanoFEVER_cosine_recall@5": 0.18, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.32, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.17161850356661693, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.26146031746031745, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.18168469434151963, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.07600000000000001, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.1, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.19, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.66, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4690858388149529, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.45969047619047615, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.5031291113830936, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.132, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.66, | |
| "eval_mldr_loss": 1.7250500917434692, | |
| "eval_mldr_runtime": 14.4371, | |
| "eval_mldr_samples_per_second": 34.633, | |
| "eval_mldr_steps_per_second": 2.217, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6482982171799028, | |
| "eval_squad_loss": 0.8224272131919861, | |
| "eval_squad_runtime": 0.9831, | |
| "eval_squad_samples_per_second": 48.823, | |
| "eval_squad_steps_per_second": 3.051, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6482982171799028, | |
| "eval_narrative_qa_loss": 2.191936492919922, | |
| "eval_narrative_qa_runtime": 1.1776, | |
| "eval_narrative_qa_samples_per_second": 253.914, | |
| "eval_narrative_qa_steps_per_second": 16.135, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6645056726094003, | |
| "grad_norm": 796.0, | |
| "learning_rate": 3.962448216468275e-05, | |
| "loss": 3.2988, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6807131280388979, | |
| "grad_norm": 9216.0, | |
| "learning_rate": 3.9079069738783484e-05, | |
| "loss": 1.3305, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6969205834683955, | |
| "grad_norm": 227.0, | |
| "learning_rate": 3.852366144750358e-05, | |
| "loss": 1.2771, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.713128038897893, | |
| "grad_norm": 1456.0, | |
| "learning_rate": 3.7958651619915495e-05, | |
| "loss": 1.0724, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.7293354943273906, | |
| "grad_norm": 704.0, | |
| "learning_rate": 3.738444140199549e-05, | |
| "loss": 1.2473, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7455429497568882, | |
| "grad_norm": 4608.0, | |
| "learning_rate": 3.680143847181783e-05, | |
| "loss": 1.4045, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7617504051863857, | |
| "grad_norm": 8384.0, | |
| "learning_rate": 3.621005675011127e-05, | |
| "loss": 1.6167, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7779578606158833, | |
| "grad_norm": 10752.0, | |
| "learning_rate": 3.5610716106383426e-05, | |
| "loss": 1.427, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7941653160453809, | |
| "grad_norm": 32512.0, | |
| "learning_rate": 3.500384206082155e-05, | |
| "loss": 1.256, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8103727714748784, | |
| "grad_norm": 2080.0, | |
| "learning_rate": 3.438986548218155e-05, | |
| "loss": 1.7336, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8103727714748784, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.22666666666666668, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.35333333333333333, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.25285734426656203, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.28679100529100526, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.22666666666666668, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.26096692405722594, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.22666666666666668, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.07600000000000001, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.17666666666666667, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.31, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.08, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.18, | |
| "eval_NanoFEVER_cosine_map@100": 0.11514089868666741, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.11641269841269843, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.08, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.11458961164975079, | |
| "eval_NanoFEVER_cosine_precision@1": 0.08, | |
| "eval_NanoFEVER_cosine_precision@5": 0.036000000000000004, | |
| "eval_NanoFEVER_cosine_recall@1": 0.07, | |
| "eval_NanoFEVER_cosine_recall@5": 0.16, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.28, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.3, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.1921273663105645, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.3077380952380952, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.28, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.202243938877232, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.28, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.07600000000000001, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.14, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.19, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.58, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4513037678024542, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.4362222222222223, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.4660672216446949, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.11600000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.58, | |
| "eval_mldr_loss": 1.2118109464645386, | |
| "eval_mldr_runtime": 14.4727, | |
| "eval_mldr_samples_per_second": 34.548, | |
| "eval_mldr_steps_per_second": 2.211, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8103727714748784, | |
| "eval_squad_loss": 0.669793426990509, | |
| "eval_squad_runtime": 0.9528, | |
| "eval_squad_samples_per_second": 50.377, | |
| "eval_squad_steps_per_second": 3.149, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8103727714748784, | |
| "eval_narrative_qa_loss": 2.0540804862976074, | |
| "eval_narrative_qa_runtime": 1.1722, | |
| "eval_narrative_qa_samples_per_second": 255.077, | |
| "eval_narrative_qa_steps_per_second": 16.209, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.826580226904376, | |
| "grad_norm": 5792.0, | |
| "learning_rate": 3.3769222281879495e-05, | |
| "loss": 0.9241, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.8427876823338736, | |
| "grad_norm": 1080.0, | |
| "learning_rate": 3.314235310450306e-05, | |
| "loss": 1.2186, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.8589951377633711, | |
| "grad_norm": 1264.0, | |
| "learning_rate": 3.250970301496237e-05, | |
| "loss": 1.3873, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.8752025931928687, | |
| "grad_norm": 308.0, | |
| "learning_rate": 3.187172118250266e-05, | |
| "loss": 1.299, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8914100486223663, | |
| "grad_norm": 5568.0, | |
| "learning_rate": 3.122886056180284e-05, | |
| "loss": 1.4101, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9076175040518638, | |
| "grad_norm": 1480.0, | |
| "learning_rate": 3.0581577571386426e-05, | |
| "loss": 1.047, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9238249594813615, | |
| "grad_norm": 34.0, | |
| "learning_rate": 2.9930331769573365e-05, | |
| "loss": 1.1558, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.940032414910859, | |
| "grad_norm": 4640.0, | |
| "learning_rate": 2.9275585528202516e-05, | |
| "loss": 1.1769, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.9562398703403565, | |
| "grad_norm": 191.0, | |
| "learning_rate": 2.8617803704356672e-05, | |
| "loss": 1.1054, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.9724473257698542, | |
| "grad_norm": 167.0, | |
| "learning_rate": 2.7957453310323073e-05, | |
| "loss": 3.6745, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9724473257698542, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.20666666666666667, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.36000000000000004, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.24725005259175806, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.2764206349206349, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.20666666666666667, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.2557813556128824, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.20666666666666667, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.076, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.16666666666666666, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.31, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.1, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.12, | |
| "eval_NanoFEVER_cosine_map@100": 0.12500495211409104, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.12576984126984128, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.1, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.10488153892673832, | |
| "eval_NanoFEVER_cosine_precision@1": 0.1, | |
| "eval_NanoFEVER_cosine_precision@5": 0.024000000000000004, | |
| "eval_NanoFEVER_cosine_recall@1": 0.09, | |
| "eval_NanoFEVER_cosine_recall@5": 0.11, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.22, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.34, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.17072760263191902, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.2686904761904762, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.22, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.19019906156669042, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.22, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.08, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.11, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.2, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.62, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.44601760302926413, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.4348015873015873, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.4722634663452185, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.124, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.3, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.62, | |
| "eval_mldr_loss": 1.1025381088256836, | |
| "eval_mldr_runtime": 14.5605, | |
| "eval_mldr_samples_per_second": 34.339, | |
| "eval_mldr_steps_per_second": 2.198, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9724473257698542, | |
| "eval_squad_loss": 0.7168652415275574, | |
| "eval_squad_runtime": 0.8907, | |
| "eval_squad_samples_per_second": 53.891, | |
| "eval_squad_steps_per_second": 3.368, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9724473257698542, | |
| "eval_narrative_qa_loss": 1.757150411605835, | |
| "eval_narrative_qa_runtime": 1.1289, | |
| "eval_narrative_qa_samples_per_second": 264.853, | |
| "eval_narrative_qa_steps_per_second": 16.83, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9886547811993517, | |
| "grad_norm": 414.0, | |
| "learning_rate": 2.7295003182023793e-05, | |
| "loss": 1.2095, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.0048622366288493, | |
| "grad_norm": 760.0, | |
| "learning_rate": 2.663092364615134e-05, | |
| "loss": 1.0888, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.0210696920583469, | |
| "grad_norm": 884.0, | |
| "learning_rate": 2.596568618624588e-05, | |
| "loss": 1.1728, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.0372771474878444, | |
| "grad_norm": 1568.0, | |
| "learning_rate": 2.529976310795108e-05, | |
| "loss": 1.2033, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.053484602917342, | |
| "grad_norm": 716.0, | |
| "learning_rate": 2.4633627203686322e-05, | |
| "loss": 1.0108, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.0696920583468394, | |
| "grad_norm": 50.5, | |
| "learning_rate": 2.3967751416973304e-05, | |
| "loss": 0.9941, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.0858995137763372, | |
| "grad_norm": 39.5, | |
| "learning_rate": 2.3302608506655286e-05, | |
| "loss": 1.1392, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.1021069692058347, | |
| "grad_norm": 1000.0, | |
| "learning_rate": 2.2638670711247605e-05, | |
| "loss": 1.083, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.1183144246353323, | |
| "grad_norm": 7456.0, | |
| "learning_rate": 2.197640941365743e-05, | |
| "loss": 1.0115, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.1345218800648298, | |
| "grad_norm": 2944.0, | |
| "learning_rate": 2.131629480651116e-05, | |
| "loss": 1.2825, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1345218800648298, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.34, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.2227183429409336, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.25063756613756616, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.23117905501265376, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.07066666666666667, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.14, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.29000000000000004, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.06, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.14, | |
| "eval_NanoFEVER_cosine_map@100": 0.10226378102099443, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.107, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.06, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.09611366515963533, | |
| "eval_NanoFEVER_cosine_precision@1": 0.06, | |
| "eval_NanoFEVER_cosine_precision@5": 0.027999999999999997, | |
| "eval_NanoFEVER_cosine_recall@1": 0.05, | |
| "eval_NanoFEVER_cosine_recall@5": 0.13, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.18, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.32, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.1431808584557489, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.2346666666666667, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.18, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.16151062083926612, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.18, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.07200000000000001, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.09, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.18, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.28, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.56, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4227103893460575, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.4102460317460318, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.28, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.4359128790390598, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.28, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.11200000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.28, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.56, | |
| "eval_mldr_loss": 1.04865562915802, | |
| "eval_mldr_runtime": 14.5246, | |
| "eval_mldr_samples_per_second": 34.424, | |
| "eval_mldr_steps_per_second": 2.203, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1345218800648298, | |
| "eval_squad_loss": 0.7077043652534485, | |
| "eval_squad_runtime": 0.907, | |
| "eval_squad_samples_per_second": 52.924, | |
| "eval_squad_steps_per_second": 3.308, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1345218800648298, | |
| "eval_narrative_qa_loss": 1.7767502069473267, | |
| "eval_narrative_qa_runtime": 1.1658, | |
| "eval_narrative_qa_samples_per_second": 256.487, | |
| "eval_narrative_qa_steps_per_second": 16.298, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1507293354943273, | |
| "grad_norm": 1976.0, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.1286, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.1669367909238249, | |
| "grad_norm": 1640.0, | |
| "learning_rate": 2.000437848076822e-05, | |
| "loss": 1.174, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.1831442463533226, | |
| "grad_norm": 3776.0, | |
| "learning_rate": 1.935350819721849e-05, | |
| "loss": 1.124, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.1993517017828201, | |
| "grad_norm": 5568.0, | |
| "learning_rate": 1.870664681290575e-05, | |
| "loss": 0.9694, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.2155591572123177, | |
| "grad_norm": 532.0, | |
| "learning_rate": 1.8064253586817816e-05, | |
| "loss": 1.165, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2317666126418152, | |
| "grad_norm": 928.0, | |
| "learning_rate": 1.7426784605637153e-05, | |
| "loss": 1.1989, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.2479740680713127, | |
| "grad_norm": 560.0, | |
| "learning_rate": 1.6794692459928223e-05, | |
| "loss": 0.9966, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.2641815235008105, | |
| "grad_norm": 123.5, | |
| "learning_rate": 1.6168425922807008e-05, | |
| "loss": 0.9768, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.280388978930308, | |
| "grad_norm": 884.0, | |
| "learning_rate": 1.5548429631320806e-05, | |
| "loss": 1.1039, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.2965964343598055, | |
| "grad_norm": 15.75, | |
| "learning_rate": 1.4935143770764601e-05, | |
| "loss": 2.7281, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2965964343598055, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.35333333333333333, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.2232392549386716, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.24966402116402117, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.23771617121243702, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.17333333333333334, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.07333333333333333, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.1366666666666667, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.31, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.06, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.16, | |
| "eval_NanoFEVER_cosine_map@100": 0.1100930297556344, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.11335714285714285, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.06, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.10785578521428746, | |
| "eval_NanoFEVER_cosine_precision@1": 0.06, | |
| "eval_NanoFEVER_cosine_precision@5": 0.032, | |
| "eval_NanoFEVER_cosine_recall@1": 0.05, | |
| "eval_NanoFEVER_cosine_recall@5": 0.15, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.28, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.14784815274029572, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.2350555555555556, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.16017374485985264, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.064, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.1, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.16, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.26, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.62, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4117765823200847, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.40057936507936504, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.26, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.445118983563171, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.26, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.124, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.26, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.62, | |
| "eval_mldr_loss": 1.0107322931289673, | |
| "eval_mldr_runtime": 14.5338, | |
| "eval_mldr_samples_per_second": 34.403, | |
| "eval_mldr_steps_per_second": 2.202, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2965964343598055, | |
| "eval_squad_loss": 0.6025272607803345, | |
| "eval_squad_runtime": 0.8024, | |
| "eval_squad_samples_per_second": 59.822, | |
| "eval_squad_steps_per_second": 3.739, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2965964343598055, | |
| "eval_narrative_qa_loss": 1.7681714296340942, | |
| "eval_narrative_qa_runtime": 1.1303, | |
| "eval_narrative_qa_samples_per_second": 264.533, | |
| "eval_narrative_qa_steps_per_second": 16.81, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.312803889789303, | |
| "grad_norm": 4288.0, | |
| "learning_rate": 1.4329003762158106e-05, | |
| "loss": 1.2271, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.3290113452188006, | |
| "grad_norm": 11840.0, | |
| "learning_rate": 1.3730439953105243e-05, | |
| "loss": 1.0453, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.3452188006482984, | |
| "grad_norm": 1440.0, | |
| "learning_rate": 1.3139877312255799e-05, | |
| "loss": 1.106, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.3614262560777957, | |
| "grad_norm": 440.0, | |
| "learning_rate": 1.2557735127585917e-05, | |
| "loss": 1.0065, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.3776337115072934, | |
| "grad_norm": 616.0, | |
| "learning_rate": 1.1984426708711804e-05, | |
| "loss": 1.124, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.393841166936791, | |
| "grad_norm": 12.5, | |
| "learning_rate": 1.1420359093447966e-05, | |
| "loss": 1.3403, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.4100486223662885, | |
| "grad_norm": 1280.0, | |
| "learning_rate": 1.0865932758818258e-05, | |
| "loss": 1.1274, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.426256077795786, | |
| "grad_norm": 186.0, | |
| "learning_rate": 1.0321541336725027e-05, | |
| "loss": 1.1335, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.4424635332252835, | |
| "grad_norm": 6656.0, | |
| "learning_rate": 9.787571334478101e-06, | |
| "loss": 1.0805, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.4586709886547813, | |
| "grad_norm": 73.5, | |
| "learning_rate": 9.26440186038211e-06, | |
| "loss": 1.1117, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4586709886547813, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.2066666666666667, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.3466666666666667, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.24109707671460848, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.27005291005291004, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.2066666666666667, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.24663303266901904, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.2066666666666667, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.07200000000000001, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.17, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.29666666666666663, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.08, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.14, | |
| "eval_NanoFEVER_cosine_map@100": 0.10717352446028915, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.1099126984126984, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.08, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.10123212623289701, | |
| "eval_NanoFEVER_cosine_precision@1": 0.08, | |
| "eval_NanoFEVER_cosine_precision@5": 0.028000000000000004, | |
| "eval_NanoFEVER_cosine_recall@1": 0.07, | |
| "eval_NanoFEVER_cosine_recall@5": 0.13, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.32, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.15637152907438528, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.2523888888888889, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.17287282180469699, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.2, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.07200000000000001, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.1, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.18, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.34, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.58, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.45974617660915107, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.44785714285714284, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.34, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.4657941499694631, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.34, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.11600000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.34, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.58, | |
| "eval_mldr_loss": 1.006673812866211, | |
| "eval_mldr_runtime": 14.487, | |
| "eval_mldr_samples_per_second": 34.514, | |
| "eval_mldr_steps_per_second": 2.209, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4586709886547813, | |
| "eval_squad_loss": 0.7755841612815857, | |
| "eval_squad_runtime": 0.8044, | |
| "eval_squad_samples_per_second": 59.675, | |
| "eval_squad_steps_per_second": 3.73, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4586709886547813, | |
| "eval_narrative_qa_loss": 1.7425211668014526, | |
| "eval_narrative_qa_runtime": 1.1372, | |
| "eval_narrative_qa_samples_per_second": 262.924, | |
| "eval_narrative_qa_steps_per_second": 16.708, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4748784440842788, | |
| "grad_norm": 712.0, | |
| "learning_rate": 8.752404354577052e-06, | |
| "loss": 1.0897, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.4910858995137763, | |
| "grad_norm": 1200.0, | |
| "learning_rate": 8.251942325322903e-06, | |
| "loss": 0.9964, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.5072933549432739, | |
| "grad_norm": 2016.0, | |
| "learning_rate": 7.763371090915932e-06, | |
| "loss": 0.954, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.5235008103727714, | |
| "grad_norm": 4512.0, | |
| "learning_rate": 7.287037527419604e-06, | |
| "loss": 1.3415, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.5397082658022692, | |
| "grad_norm": 438.0, | |
| "learning_rate": 6.823279822389278e-06, | |
| "loss": 0.9926, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.5559157212317665, | |
| "grad_norm": 158.0, | |
| "learning_rate": 6.372427234765688e-06, | |
| "loss": 1.1362, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.5721231766612642, | |
| "grad_norm": 163.0, | |
| "learning_rate": 5.934799861107507e-06, | |
| "loss": 1.1178, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.5883306320907618, | |
| "grad_norm": 1216.0, | |
| "learning_rate": 5.5107084083289576e-06, | |
| "loss": 0.8676, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.6045380875202593, | |
| "grad_norm": 37632.0, | |
| "learning_rate": 5.1004539731040245e-06, | |
| "loss": 1.1794, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.620745542949757, | |
| "grad_norm": 268.0, | |
| "learning_rate": 4.704327828093641e-06, | |
| "loss": 1.0575, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.620745542949757, | |
| "eval_NanoBEIR_mean_cosine_accuracy@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_accuracy@5": 0.3333333333333333, | |
| "eval_NanoBEIR_mean_cosine_map@100": 0.23251229334833623, | |
| "eval_NanoBEIR_mean_cosine_mrr@10": 0.250457671957672, | |
| "eval_NanoBEIR_mean_cosine_ndcg@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_ndcg@5": 0.23607659211604262, | |
| "eval_NanoBEIR_mean_cosine_precision@1": 0.18666666666666668, | |
| "eval_NanoBEIR_mean_cosine_precision@5": 0.06933333333333334, | |
| "eval_NanoBEIR_mean_cosine_recall@1": 0.15666666666666668, | |
| "eval_NanoBEIR_mean_cosine_recall@5": 0.29, | |
| "eval_NanoFEVER_cosine_accuracy@1": 0.08, | |
| "eval_NanoFEVER_cosine_accuracy@5": 0.14, | |
| "eval_NanoFEVER_cosine_map@100": 0.10995542705065009, | |
| "eval_NanoFEVER_cosine_mrr@10": 0.11060317460317462, | |
| "eval_NanoFEVER_cosine_ndcg@1": 0.08, | |
| "eval_NanoFEVER_cosine_ndcg@5": 0.10261859507142916, | |
| "eval_NanoFEVER_cosine_precision@1": 0.08, | |
| "eval_NanoFEVER_cosine_precision@5": 0.027999999999999997, | |
| "eval_NanoFEVER_cosine_recall@1": 0.07, | |
| "eval_NanoFEVER_cosine_recall@5": 0.13, | |
| "eval_NanoHotpotQA_cosine_accuracy@1": 0.16, | |
| "eval_NanoHotpotQA_cosine_accuracy@5": 0.28, | |
| "eval_NanoHotpotQA_cosine_map@100": 0.14864943925499563, | |
| "eval_NanoHotpotQA_cosine_mrr@10": 0.21274603174603177, | |
| "eval_NanoHotpotQA_cosine_ndcg@1": 0.16, | |
| "eval_NanoHotpotQA_cosine_ndcg@5": 0.15294562020041982, | |
| "eval_NanoHotpotQA_cosine_precision@1": 0.16, | |
| "eval_NanoHotpotQA_cosine_precision@5": 0.064, | |
| "eval_NanoHotpotQA_cosine_recall@1": 0.08, | |
| "eval_NanoHotpotQA_cosine_recall@5": 0.16, | |
| "eval_NanoMSMARCO_cosine_accuracy@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_accuracy@5": 0.58, | |
| "eval_NanoMSMARCO_cosine_map@100": 0.4389320137393629, | |
| "eval_NanoMSMARCO_cosine_mrr@10": 0.42802380952380953, | |
| "eval_NanoMSMARCO_cosine_ndcg@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_ndcg@5": 0.4526655610762789, | |
| "eval_NanoMSMARCO_cosine_precision@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_precision@5": 0.11600000000000002, | |
| "eval_NanoMSMARCO_cosine_recall@1": 0.32, | |
| "eval_NanoMSMARCO_cosine_recall@5": 0.58, | |
| "eval_mldr_loss": 1.0088629722595215, | |
| "eval_mldr_runtime": 14.5077, | |
| "eval_mldr_samples_per_second": 34.464, | |
| "eval_mldr_steps_per_second": 2.206, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.620745542949757, | |
| "eval_squad_loss": 0.6442683339118958, | |
| "eval_squad_runtime": 0.8625, | |
| "eval_squad_samples_per_second": 55.652, | |
| "eval_squad_steps_per_second": 3.478, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.620745542949757, | |
| "eval_narrative_qa_loss": 1.7451977729797363, | |
| "eval_narrative_qa_runtime": 1.1443, | |
| "eval_narrative_qa_samples_per_second": 261.298, | |
| "eval_narrative_qa_steps_per_second": 16.604, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1234, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |