| { |
| "config": { |
| "model": "LiquidAI/LFM2.5-Encoder-350M-Spellchecker", |
| "output_dir": "/tmp/minima-spellcheck", |
| "output_repo": "ProCreations/minima-spellcheck", |
| "dataset": "HuggingFaceFW/fineweb-edu", |
| "dataset_config": "sample-10BT", |
| "sequence_length": 128, |
| "batch_size": 2, |
| "gradient_accumulation": 4, |
| "steps": 1000, |
| "learning_rate": 5e-05, |
| "group_size": 32, |
| "recovery_rank": 128, |
| "topk": 16, |
| "eval_batches": 8, |
| "seed": 77, |
| "log_every": 50 |
| }, |
| "duration_seconds": 394.26561307907104, |
| "trainable_parameters": 57020424, |
| "tagger_label_top1_candidate_agreement": 0.9951171875, |
| "detection_top1_agreement": 0.994140625, |
| "correction_exact_agreement": 0.7, |
| "correction_evaluation_examples": 20, |
| "example_exact_agreement": 1.0, |
| "examples": [ |
| { |
| "input": "She go to school every day .", |
| "teacher": "She goes to school every day .", |
| "minima": "She goes to school every day ." |
| }, |
| { |
| "input": "I has went to the stor yesterday .", |
| "teacher": "I went to the store yesterday .", |
| "minima": "I went to the store yesterday ." |
| }, |
| { |
| "input": "Their are many reason to study hard .", |
| "teacher": "There are many reasons to study hard .", |
| "minima": "There are many reasons to study hard ." |
| }, |
| { |
| "input": "That 's a fair point , let 's discuss it tomorrow .", |
| "teacher": "That 's a fair point , let 's discuss it tomorrow .", |
| "minima": "That 's a fair point , let 's discuss it tomorrow ." |
| } |
| ], |
| "history": [ |
| { |
| "step": 1, |
| "elapsed_s": 9.728960990905762, |
| "loss": 2.4879839420318604, |
| "label": 1.3699460625648499, |
| "detect": 0.3631512150168419, |
| "hidden": 0.4682311415672302 |
| }, |
| { |
| "step": 51, |
| "elapsed_s": 47.39888334274292, |
| "loss": 0.5646287947893143, |
| "label": 0.21763740107417107, |
| "detect": 0.027677168138325214, |
| "hidden": 0.16657640412449837 |
| }, |
| { |
| "step": 101, |
| "elapsed_s": 64.47661709785461, |
| "loss": 0.433330237865448, |
| "label": 0.14443234354257584, |
| "detect": 0.012719596037641168, |
| "hidden": 0.1412690468132496 |
| }, |
| { |
| "step": 151, |
| "elapsed_s": 81.66340661048889, |
| "loss": 0.374489426612854, |
| "label": 0.12132998369634151, |
| "detect": 0.0135470035020262, |
| "hidden": 0.12319297343492508 |
| }, |
| { |
| "step": 201, |
| "elapsed_s": 98.94163203239441, |
| "loss": 0.38375477492809296, |
| "label": 0.12871305830776691, |
| "detect": 0.015309339622035623, |
| "hidden": 0.12369352579116821 |
| }, |
| { |
| "step": 251, |
| "elapsed_s": 116.117196559906, |
| "loss": 0.4129961282014847, |
| "label": 0.16921869292855263, |
| "detect": 0.018067472614347935, |
| "hidden": 0.11737184971570969 |
| }, |
| { |
| "step": 301, |
| "elapsed_s": 133.41278839111328, |
| "loss": 0.3421344459056854, |
| "label": 0.12530853413045406, |
| "detect": 0.008865324780344963, |
| "hidden": 0.10619662329554558 |
| }, |
| { |
| "step": 351, |
| "elapsed_s": 150.6108753681183, |
| "loss": 0.4015732929110527, |
| "label": 0.15202416479587555, |
| "detect": 0.021410188637673855, |
| "hidden": 0.11942201666533947 |
| }, |
| { |
| "step": 401, |
| "elapsed_s": 167.74786019325256, |
| "loss": 0.3250713273882866, |
| "label": 0.11101343296468258, |
| "detect": 0.014453576179221272, |
| "hidden": 0.10341555811464787 |
| }, |
| { |
| "step": 451, |
| "elapsed_s": 184.91804909706116, |
| "loss": 0.30887674540281296, |
| "label": 0.1138718631118536, |
| "detect": 0.009780762484297156, |
| "hidden": 0.09505725093185902 |
| }, |
| { |
| "step": 501, |
| "elapsed_s": 201.9468400478363, |
| "loss": 0.2911076098680496, |
| "label": 0.093521011993289, |
| "detect": 0.011887337197549641, |
| "hidden": 0.09582146629691124 |
| }, |
| { |
| "step": 551, |
| "elapsed_s": 218.95755100250244, |
| "loss": 0.3279508873820305, |
| "label": 0.12233076430857182, |
| "detect": 0.012748322915285826, |
| "hidden": 0.09962298162281513 |
| }, |
| { |
| "step": 601, |
| "elapsed_s": 235.95491790771484, |
| "loss": 0.2837226167321205, |
| "label": 0.08802595920860767, |
| "detect": 0.006528681959025562, |
| "hidden": 0.09621615894138813 |
| }, |
| { |
| "step": 651, |
| "elapsed_s": 252.9777970314026, |
| "loss": 0.29664454981684685, |
| "label": 0.1072889119386673, |
| "detect": 0.010023394948802888, |
| "hidden": 0.09217197075486183 |
| }, |
| { |
| "step": 701, |
| "elapsed_s": 270.00525283813477, |
| "loss": 0.31047341972589493, |
| "label": 0.11355133540928364, |
| "detect": 0.01577813387848437, |
| "hidden": 0.09451650828123093 |
| }, |
| { |
| "step": 751, |
| "elapsed_s": 287.11339259147644, |
| "loss": 0.2924303859472275, |
| "label": 0.09383961372077465, |
| "detect": 0.008933465695008636, |
| "hidden": 0.09706202149391174 |
| }, |
| { |
| "step": 801, |
| "elapsed_s": 304.38399958610535, |
| "loss": 0.26952020823955536, |
| "label": 0.08447664976119995, |
| "detect": 0.01343391730915755, |
| "hidden": 0.08916330337524414 |
| }, |
| { |
| "step": 851, |
| "elapsed_s": 321.6072781085968, |
| "loss": 0.27088215947151184, |
| "label": 0.08796967100352049, |
| "detect": 0.008289208286441863, |
| "hidden": 0.08938394114375114 |
| }, |
| { |
| "step": 901, |
| "elapsed_s": 338.75015807151794, |
| "loss": 0.2826335057616234, |
| "label": 0.09538542665541172, |
| "detect": 0.010135660180822015, |
| "hidden": 0.0910901241004467 |
| }, |
| { |
| "step": 951, |
| "elapsed_s": 355.75725746154785, |
| "loss": 0.2881084829568863, |
| "label": 0.09700636379420757, |
| "detect": 0.014985204674303532, |
| "hidden": 0.0918047558516264 |
| }, |
| { |
| "step": 1000, |
| "elapsed_s": 372.5106551647186, |
| "loss": 0.3113468959927559, |
| "label": 0.1258483398705721, |
| "detect": 0.012814029469154775, |
| "hidden": 0.08954576589167118 |
| } |
| ] |
| } |
|
|