| { |
| "date": "2026-10-01", |
| "model": "d0rj/diffusion-51M-base", |
| "protocol": "Full selected splits; lm-eval 0.4.12; seed 1234; BF16 on RTX 5070 Ti;\ncontext cap 2048 (ArithMark 1024), TF32 disabled, no chat template and no added\nfew-shot examples. TruthfulQA retains the harness's fixed six-QA preamble.\nArithMark and BananaMind normalize by continuation token count; ordinary harness\nacc_norm uses its own length normalization. BananaMind is raw accuracy, not Elo.\nSciQ includes the support passage. Balanced COPA uses the mirrored 1000-item\ntrain-named evaluation split; cRia's split was inferred, not confirmed.\nMMLU scores full answer continuations across 57 subjects, weighted by item count;\nBLiMP averages 67 equal-sized minimal-pair subsets. Standard errors are retained\nfrom each evaluator. Wilson intervals are reported only where the runner logged\nbinary item accuracy; MC2 is probability mass, not binary accuracy. These\nintervals do not model dependence between paired/templated examples or training\nseed variation. UL2, PrefixLM and experimental diffusion PLL use their documented\nconditional scoring protocols; PLL exposes the other answer tokens and is not\nautoregressive likelihood. cRia's published scores used a different precision\nand benchmark-adapted checkpoint; this completes our comparison coverage, not\nan independent reproduction of cRia or an official leaderboard submission.", |
| "results": { |
| "checkpoint": "/mnt/d/Projects/tiny_llm/runs/eval/diffusion-v2-step15000-20260923", |
| "checkpoint_sha256": { |
| "config.json": "8e3daed6864d9b1f39be95c7b311a93c47efd4e76ee33fb4dbeede6d4701343d", |
| "model.safetensors": "d0d24e0cc492cd5c9a613710c454af9a6e49265d2bae7abb69a673a53efe2bfb" |
| }, |
| "core": "/mnt/d/Projects/tiny_llm/runs/eval/diffusion-v2-core-full-20260923", |
| "training_run": "diffusion-gpu-v2", |
| "training_step": 15000, |
| "hf": "d0rj/diffusion-51M-base", |
| "tokenizer": "/mnt/d/Projects/tiny_llm/tokenizer", |
| "tb": "runs/diffusion-gpu-v2/tb", |
| "protocol": "EXPERIMENTAL: continuation PLL; single-mask left-to-right greedy generation; NOT AR likelihood", |
| "kind": "diffusion", |
| "tasks": { |
| "arithmark3": { |
| "label": "ArithMark-3", |
| "dataset": "AxiomicLabs/Arithmark-3.0", |
| "config": "default", |
| "split": "train", |
| "samples": 1000, |
| "shots": 0, |
| "primary": "acc_norm", |
| "metrics": { |
| "acc": { |
| "value": 0.332, |
| "stderr": 0.014899597242811565, |
| "ci95": [ |
| 0.30350363489239063, |
| 0.36178215595875596 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| }, |
| "acc_norm": { |
| "value": 0.332, |
| "stderr": 0.014899597242811565, |
| "ci95": [ |
| 0.30350363489239063, |
| 0.36178215595875596 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/arithmark3.json", |
| "result_sha256": "68430957620991dec536d3717f684c1b92d6921b9e324940f542ceb62109ec0f", |
| "categories": { |
| "elementary_school_math_continuation::addition::grades_1_2::easy": { |
| "count": 128, |
| "acc_norm,none": 0.21875, |
| "acc_norm_stderr,none": 0.03668319762764659 |
| }, |
| "elementary_school_math_continuation::comparison::grades_2_3::medium": { |
| "count": 44, |
| "acc_norm,none": 0.18181818181818182, |
| "acc_norm_stderr,none": 0.05881787629278459 |
| }, |
| "elementary_school_math_continuation::comparison_difference::grades_2_3::medium": { |
| "count": 48, |
| "acc_norm,none": 0.2708333333333333, |
| "acc_norm_stderr,none": 0.06482097094483913 |
| }, |
| "elementary_school_math_continuation::data::grades_2_3::easy": { |
| "count": 43, |
| "acc_norm,none": 0.16279069767441862, |
| "acc_norm_stderr,none": 0.056964877739143674 |
| }, |
| "elementary_school_math_continuation::division::grades_3_4::medium": { |
| "count": 54, |
| "acc_norm,none": 0.37037037037037035, |
| "acc_norm_stderr,none": 0.06633194954624339 |
| }, |
| "elementary_school_math_continuation::fractions_counting::grades_3_4::medium": { |
| "count": 50, |
| "acc_norm,none": 0.08, |
| "acc_norm_stderr,none": 0.03875617133214439 |
| }, |
| "elementary_school_math_continuation::geometry_area::grades_4_5::medium": { |
| "count": 52, |
| "acc_norm,none": 0.46153846153846156, |
| "acc_norm_stderr,none": 0.06980655484407924 |
| }, |
| "elementary_school_math_continuation::geometry_perimeter::grades_4_5::medium": { |
| "count": 45, |
| "acc_norm,none": 0.4666666666666667, |
| "acc_norm_stderr,none": 0.0752101433090355 |
| }, |
| "elementary_school_math_continuation::measurement::grades_2_3::easy": { |
| "count": 76, |
| "acc_norm,none": 0.32894736842105265, |
| "acc_norm_stderr,none": 0.05425138981075869 |
| }, |
| "elementary_school_math_continuation::money::grades_3_4::medium": { |
| "count": 64, |
| "acc_norm,none": 0.375, |
| "acc_norm_stderr,none": 0.060993754559283325 |
| }, |
| "elementary_school_math_continuation::multiplication::grades_3_4::medium": { |
| "count": 74, |
| "acc_norm,none": 0.5675675675675675, |
| "acc_norm_stderr,none": 0.057983774751431 |
| }, |
| "elementary_school_math_continuation::patterns::grades_3_4::medium": { |
| "count": 53, |
| "acc_norm,none": 0.16981132075471697, |
| "acc_norm_stderr,none": 0.05206789873629056 |
| }, |
| "elementary_school_math_continuation::subtraction::grades_1_2::easy": { |
| "count": 117, |
| "acc_norm,none": 0.23931623931623933, |
| "acc_norm_stderr,none": 0.03961495460787807 |
| }, |
| "elementary_school_math_continuation::time::grades_2_3::easy": { |
| "count": 55, |
| "acc_norm,none": 0.9818181818181818, |
| "acc_norm_stderr,none": 0.018181818181818184 |
| }, |
| "elementary_school_math_continuation::two_step_add_subtract::grades_2_3::medium": { |
| "count": 46, |
| "acc_norm,none": 0.21739130434782608, |
| "acc_norm_stderr,none": 0.06148754619013457 |
| }, |
| "elementary_school_math_continuation::two_step_addition::grades_2_3::medium": { |
| "count": 19, |
| "acc_norm,none": 0.3157894736842105, |
| "acc_norm_stderr,none": 0.10956136839295436 |
| }, |
| "elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium": { |
| "count": 32, |
| "acc_norm,none": 0.28125, |
| "acc_norm_stderr,none": 0.08075219711382271 |
| } |
| } |
| }, |
| "balanced_copa": { |
| "label": "Balanced COPA", |
| "dataset": "pkavumba/balanced-copa", |
| "config": "default", |
| "split": "train", |
| "samples": 1000, |
| "shots": 0, |
| "primary": "acc", |
| "metrics": { |
| "acc": { |
| "value": 0.522, |
| "stderr": 0.01580397942816194, |
| "ci95": [ |
| 0.4910152521271681, |
| 0.5528163704994674 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/balanced_copa.json", |
| "result_sha256": "1375e8cff071a9464a7c0eded68c0afd7b944bb716d680493dfbe3301963b482", |
| "categories": {} |
| }, |
| "commonsense_qa": { |
| "label": "CommonsenseQA", |
| "dataset": "tau/commonsense_qa", |
| "config": "default", |
| "split": "validation", |
| "samples": 1221, |
| "shots": 0, |
| "primary": "acc", |
| "metrics": { |
| "acc": { |
| "value": 0.21457821457821458, |
| "stderr": 0.011753423094216953, |
| "ci95": [ |
| 0.19246524693116476, |
| 0.2384815135817527 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/commonsense_qa.json", |
| "result_sha256": "28a06065e6b827e085a8a03ce1bba594ae9edf1ae0cc7d5814b01be992f064bb", |
| "categories": {} |
| }, |
| "sciq": { |
| "label": "SciQ (with support)", |
| "dataset": "allenai/sciq", |
| "config": "default", |
| "split": "test", |
| "samples": 1000, |
| "shots": 0, |
| "primary": "acc_norm", |
| "metrics": { |
| "acc": { |
| "value": 0.723, |
| "stderr": 0.014158794845306273, |
| "ci95": [ |
| 0.6944497560637553, |
| 0.7498435096516871 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| }, |
| "acc_norm": { |
| "value": 0.688, |
| "stderr": 0.014658474370509057, |
| "ci95": [ |
| 0.6586108249018996, |
| 0.7159503139075316 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/sciq.json", |
| "result_sha256": "0f1d7377097498ae7ea47e0c1b70941aedbcea59f3e2b3ca0c0101083e6af4ee", |
| "categories": {} |
| }, |
| "truthfulqa_mc2": { |
| "label": "TruthfulQA MC2", |
| "dataset": "truthfulqa/truthful_qa", |
| "config": "multiple_choice", |
| "split": "validation", |
| "samples": 817, |
| "shots": 0, |
| "primary": "acc", |
| "metrics": { |
| "acc": { |
| "value": 0.45836334441298165, |
| "stderr": 0.0158524239677764 |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/truthfulqa_mc2.json", |
| "result_sha256": "bfc8d62f669c951f26cc2d300dbc4bd7c7787001573e0fc310c2db27c9405541", |
| "categories": {} |
| }, |
| "bananamind_base": { |
| "label": "BananaMind Base 1.1", |
| "dataset": "BananaMind/BananaMind-Base-Bench-1.1", |
| "config": "default", |
| "split": "test", |
| "samples": 350, |
| "shots": 0, |
| "primary": "raw_accuracy", |
| "metrics": { |
| "raw_accuracy": { |
| "value": 0.39714285714285713, |
| "stderr": 0.02619195222772307, |
| "ci95": [ |
| 0.34726441786546686, |
| 0.4492546213676228 |
| ], |
| "ci_method": "Wilson 95%; item independence approximation" |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/bananamind_base.json", |
| "result_sha256": "5d9eb7e2d3daa94d14dd7d35a35003c17ab1f376931e739ca027198bb566dcb7", |
| "categories": { |
| "code_completion": { |
| "count": 50, |
| "raw_accuracy,none": 0.28, |
| "raw_accuracy_stderr,none": 0.06414269805898186 |
| }, |
| "commonsense": { |
| "count": 50, |
| "raw_accuracy,none": 0.26, |
| "raw_accuracy_stderr,none": 0.06266203485560375 |
| }, |
| "context_tracking": { |
| "count": 50, |
| "raw_accuracy,none": 0.48, |
| "raw_accuracy_stderr,none": 0.07137140569598169 |
| }, |
| "language_completion": { |
| "count": 50, |
| "raw_accuracy,none": 0.4, |
| "raw_accuracy_stderr,none": 0.06998542122237651 |
| }, |
| "logical_reasoning": { |
| "count": 50, |
| "raw_accuracy,none": 0.44, |
| "raw_accuracy_stderr,none": 0.07091242083423346 |
| }, |
| "quantitative": { |
| "count": 50, |
| "raw_accuracy,none": 0.36, |
| "raw_accuracy_stderr,none": 0.06857142857142857 |
| }, |
| "world_knowledge": { |
| "count": 50, |
| "raw_accuracy,none": 0.56, |
| "raw_accuracy_stderr,none": 0.07091242083423346 |
| } |
| } |
| }, |
| "mmlu_continuation": { |
| "label": "MMLU continuation", |
| "dataset": "cais/mmlu", |
| "config": "57 subjects", |
| "split": "test", |
| "samples": 14042, |
| "shots": 0, |
| "primary": "acc", |
| "metrics": { |
| "acc": { |
| "value": 0.24483691781797465, |
| "stderr": 0.0036208715458871405 |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/mmlu_continuation.json", |
| "result_sha256": "423e99cb8501f47e672d81cfa27934e7e8986de4b85e52c5248b7ac37a01370a", |
| "categories": {} |
| }, |
| "blimp": { |
| "label": "BLiMP", |
| "dataset": "nyu-mll/blimp", |
| "config": "67 minimal-pair subsets", |
| "split": "train", |
| "samples": 67000, |
| "shots": 0, |
| "primary": "acc", |
| "metrics": { |
| "acc": { |
| "value": 0.6420746268656716, |
| "stderr": 0.0016890890018278603 |
| } |
| }, |
| "result_path": "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001/blimp.json", |
| "result_sha256": "585fbe3ad36be3a0616a31f078a27c405f136571da4859f6f6cd216cad0b0f09", |
| "categories": {} |
| } |
| }, |
| "batch_size": 1, |
| "label": "Diffusion v2 (PLL)", |
| "manifests": { |
| "runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001": { |
| "checkpoint": "/mnt/d/Projects/tiny_llm/runs/eval/diffusion-v2-step15000-20260923", |
| "local_model": "diffusion", |
| "kind": "diffusion", |
| "tokenizer": "/mnt/d/Projects/tiny_llm/tokenizer", |
| "revision": "main", |
| "tokenizer_revision": null, |
| "protocol": "EXPERIMENTAL: continuation PLL; single-mask left-to-right greedy generation; NOT AR likelihood", |
| "harness_version": "0.4.12", |
| "smoke_only": false, |
| "tasks": [ |
| { |
| "task": "arithmark3", |
| "shots": 0, |
| "description": "ArithMark-3.0 train-named evaluation split; token-normalized acc_norm", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "token_continuation" |
| }, |
| { |
| "task": "balanced_copa", |
| "shots": 0, |
| "description": "Balanced COPA mirrored 1000-item train split; acc; cRia split inferred", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| }, |
| { |
| "task": "commonsense_qa", |
| "shots": 0, |
| "description": "CommonsenseQA; acc", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| }, |
| { |
| "task": "sciq", |
| "shots": 0, |
| "description": "SciQ test with support passage; acc and acc_norm", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| }, |
| { |
| "task": "truthfulqa_mc2", |
| "shots": 0, |
| "description": "TruthfulQA multiple choice; mc2", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| }, |
| { |
| "task": "bananamind_base", |
| "shots": 0, |
| "description": "BananaMind Base Bench 1.1 test; mean-token raw accuracy; gated", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "token_continuation" |
| }, |
| { |
| "task": "mmlu_continuation", |
| "shots": 0, |
| "description": "MMLU cloze: full answer continuations; acc", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| }, |
| { |
| "task": "blimp", |
| "shots": 0, |
| "description": "Full BLiMP group; grammatical minimal pairs", |
| "unsafe_code": false, |
| "rolling": false, |
| "engine": "lm_eval" |
| } |
| ], |
| "options": { |
| "checkpoint": "/mnt/d/Projects/tiny_llm/runs/eval/diffusion-v2-step15000-20260923", |
| "model": null, |
| "tokenizer": "/mnt/d/Projects/tiny_llm/tokenizer", |
| "revision": "main", |
| "tokenizer_revision": null, |
| "suite": "core", |
| "tasks": "arithmark3,balanced_copa,commonsense_qa,sciq,truthfulqa_mc2,bananamind_base,mmlu_continuation,blimp", |
| "num_fewshot": null, |
| "device": "cuda:0", |
| "dtype": "bfloat16", |
| "allow_tf32": false, |
| "batch_size": 1, |
| "max_length": 2048, |
| "max_gen_toks": null, |
| "limit": null, |
| "seed": 1234, |
| "threads": 4, |
| "bootstrap_iters": 1000, |
| "trust_remote_code": false, |
| "allow_code_execution": false, |
| "apply_chat_template": false, |
| "log_samples": false, |
| "output": "/mnt/d/Projects/tiny_llm/runs/eval/cria-backfill-20261001/diffusion-v2/attempt-001", |
| "tb_dir": "runs/diffusion-gpu-v2/tb", |
| "training_step": 15000, |
| "list": false, |
| "dry_run": false |
| }, |
| "checkpoint_stamp": { |
| "model.safetensors": [ |
| 205580928, |
| 1790122976166062900 |
| ], |
| "config.json": [ |
| 677, |
| 1790122975189825600 |
| ] |
| }, |
| "checkpoint_sha256": { |
| "config.json": "8e3daed6864d9b1f39be95c7b311a93c47efd4e76ee33fb4dbeede6d4701343d", |
| "model.safetensors": "d0d24e0cc492cd5c9a613710c454af9a6e49265d2bae7abb69a673a53efe2bfb" |
| }, |
| "source_sha256": { |
| "src/tiny_llm/__init__.py": "8c7f075d51006a53fe1d645ef5a84d65db58ff4251e7544074fe29ee7549c14c", |
| "src/tiny_llm/benchmarks.py": "5d9f461366389bd1f926c6a6468192f3e2313c87caf5b1de27f407d563a858a4", |
| "src/tiny_llm/checkpoint.py": "ac9df5f8c981a47eb1f1e59d0687ff41a1f6c38ca74ced50f71ac446a58d6ef5", |
| "src/tiny_llm/comet_sync.py": "ea802bfb58dacad66f83b9e5ac5ccb330c190204a81c90c9e14aafe44c893c37", |
| "src/tiny_llm/comet_views.py": "3e03f4ae69f8edc2eaf5f946f5cc90f373ca1b9cc91f9dd2e9adae6d6a672689", |
| "src/tiny_llm/continuation.py": "4f4143c828d4f94a851bcbf8a466d3e1b3e9eaec1027970a426208cc0ff4f4db", |
| "src/tiny_llm/data.py": "1a52e38afef9af32463779cc2f471d9b7174a7c4c73753e74f485d526d3c6733", |
| "src/tiny_llm/diffusion_diagnostics.py": "2ae667a076b1d3f057a29d20f00559068eb73a073b3f99030e1d5b4c2f478fa5", |
| "src/tiny_llm/evaluation/__init__.py": "e49e80f7be3ab6e200f0db14b028e465de9caa8f6f705430eabce0c41b8322ba", |
| "src/tiny_llm/evaluation/adapters.py": "8e64e5542bbcb0f9ad448b320e712916a9559823987e298b22321a8d8e2f727f", |
| "src/tiny_llm/evaluation/catalog.py": "2fd29084ec780539023a38bcbfd7302931fa6be48624e227875f13518a1b986f", |
| "src/tiny_llm/evaluation/cli.py": "a322e39bf58ba52132249dae524e9772c3a8e1a5ba388f5c0267f8dc1ea18882", |
| "src/tiny_llm/evaluation/tasks/balanced_copa.yaml": "f47baa6a713e15f7a5600ed257f316f8c54252bf4977bbc716a6da7aa580f05d", |
| "src/tiny_llm/evaluation/tasks/balanced_copa_test.yaml": "dce21ad70b395d8d576c584be1b87ebae846e360e59878030645215a387b28e6", |
| "src/tiny_llm/evaluation/telemetry.py": "f4dad30574753739ea10ba71ef9a7ed9d9118a89a8202dee224961be27f13a48", |
| "src/tiny_llm/evaluation/token_continuation.py": "e8d88cd4aca51ca2fda07f2b799a54612d0e2964a98e05824a135bea1a533550", |
| "src/tiny_llm/losses.py": "592a05287369f6e5077b3faca404a9392de4772cf06389fb8aa7e36924e74f86", |
| "src/tiny_llm/models/__init__.py": "b94ab167b6eb9f9b455823eea382e67b5ac2cc7f9ef5f6de183a09b9650b9718", |
| "src/tiny_llm/models/diffusion/__init__.py": "c92eaf5986d604d8aaf1894a862a217cefa32ab25359175b88110e45a71c1b25", |
| "src/tiny_llm/models/diffusion/configuration_diffusion_lm.py": "3b303aba60f0a831c966f5cfa7f5441daa2584814e335646195bd6ceecfa6d0c", |
| "src/tiny_llm/models/diffusion/modeling_diffusion_lm.py": "4fc21c2aed1d5a231afe6f7634d9d8c3682b4e2abfc8fecef27b52136de77b2f", |
| "src/tiny_llm/models/looped/__init__.py": "4b46b426021c59ed313cb2f49afc1926ad2793d771ef90117e3fef0462957cff", |
| "src/tiny_llm/models/looped/configuration_looped_lm.py": "b8d99c52b98e0a97a4cae76ca4c20affd8f4d019ec38e0a9beefd5fad864dc25", |
| "src/tiny_llm/models/looped/modeling_looped_lm.py": "9bdb3c5e3aaae663d09e98055f95df1adbc42e0a4bf5dffb7f6c39242af41b9e", |
| "src/tiny_llm/models/moe_t5/__init__.py": "18e33fe38b2049521f573512ac8796afa209e1b25c2052784e1c927a7160d66c", |
| "src/tiny_llm/models/moe_t5/configuration_alicet5_moe.py": "4647b1e9ab9ce07887bffc1abc3beb3e691962a380b4cc237aae376490bb56b7", |
| "src/tiny_llm/models/moe_t5/modeling_alicet5_moe.py": "5f899c4726b120eed6c9bf6258ad319bb1ccea734645f27e7e6ccbc5057222ff", |
| "src/tiny_llm/models/prefixlm/__init__.py": "c94c3c861acba8d2147a3a6c46bccfc71ac8adc8d70757bede009598029d31ea", |
| "src/tiny_llm/models/prefixlm/configuration_prefix_lm.py": "20b3a62ae3628b5a64566ef19d2947c90f10c6159daad00dd13d18fc5910c42c", |
| "src/tiny_llm/models/prefixlm/modeling_prefix_lm.py": "f62a38211df175696f64945173d1e41d9f7fcf6152a7648020da8c91788e98fd", |
| "src/tiny_llm/models/q50m/__init__.py": "9d0dfec6fa0cfa6967e2f17a73d8fbeca69ba93b4a93651eda4d95626d5f5eb5", |
| "src/tiny_llm/models/q50m/configuration_q50m.py": "796806b2d5d982b633796bc7de53c3073b1b6e52d259f44ef798b63f55eaea92", |
| "src/tiny_llm/models/q50m/modeling_q50m.py": "b68abce1f19a36deb39c4ac4e520ffaab0c879d8a7cc2abef35f014bacc011f2", |
| "src/tiny_llm/models/qmod/__init__.py": "e8c3e10710ac1cceac72ea329db480e680e8bbaae2a92ec26c36c92a778552ad", |
| "src/tiny_llm/models/qmod/configuration_qmod.py": "0dfc9a4540697a9b15e423b9592df9fc435dcfc437ddad20dc6b6377ebebea40", |
| "src/tiny_llm/models/qmod/modeling_qmod.py": "3831e34a86309b91a7a0787b9471eaceb91aa4c267f8030eb2e6722a0fcab6f8", |
| "src/tiny_llm/models/qmoe/__init__.py": "cfb2bb24bc2f3d5b72b3890a393b0386cb25f127e375f5454255478a36dec32f", |
| "src/tiny_llm/models/qmoe/configuration_qmoe.py": "bba7c9ac16f444146a78c99e7a26b4a11864b215d2b673bf658d9930f2f4de2f", |
| "src/tiny_llm/models/qmoe/modeling_qmoe.py": "1d713e5e287f46dbf13b7cf9c562b4539efbf6231f7ceaa261f8cea1acd2768c", |
| "src/tiny_llm/models/qt5/__init__.py": "df943b42a9f339e006687bcd30a666838c1dec15c79751b944e5f7a2bc69186c", |
| "src/tiny_llm/models/qt5/configuration_qt5.py": "59709fe2c9fe08ccecea6633b29b3b47170fd30abcbee37952507d676446cd44", |
| "src/tiny_llm/models/qt5/modeling_qt5.py": "6269829cca6631824622df034ff7b544dbbe47790556c60b26e356b5279cceff", |
| "src/tiny_llm/models/t5/__init__.py": "e4d8838253800edbcac1f9a9b6dc875bcb3ffa2962a0c9044d066448ea10488e", |
| "src/tiny_llm/models/t5/configuration_alicet5.py": "81b1f594139f201af9afb25ba58df862432c15dedf57f6e340ee2af80ed0d44f", |
| "src/tiny_llm/models/t5/modeling_alicet5.py": "bd52ce0022668f2defb57770df229f148e070ea73becca9a2e681325114bfc20", |
| "src/tiny_llm/models/zarya/__init__.py": "651b19cdba422680ff16bf2b2f63be0c3e9045c44c410fb17dbf51df2a23bd55", |
| "src/tiny_llm/models/zarya/configuration_zarya_lm.py": "42a3e7a1b11b687bac34e00f9cd3cb59d1cd573eec7492e1c55834662d37ed3f", |
| "src/tiny_llm/models/zarya/modeling_zarya_lm.py": "73ab033860288f7367d5b912c79de908fd81a4e88f369a658b96dd2bcfcd7bfe", |
| "src/tiny_llm/telemetry.py": "7ee8f14f4545bfa0e5ac632440d49c7f86fd6812a16429f862ebe3cfe09e3545", |
| "src/tiny_llm/ul2.py": "f4d948d89d5e106965612971dd473a54b57719f811761f18457c6221d763b876", |
| "src/tiny_llm/zarya.py": "8b5084ab5086c75745115d3a9ba4f2fa9cfda8ad713f65d3f8a22e77db8c0941" |
| }, |
| "packages": { |
| "lm_eval": "0.4.12", |
| "torch": "2.11.0+cu128", |
| "transformers": "5.17.0", |
| "datasets": "5.0.1", |
| "accelerate": "1.15.0" |
| }, |
| "git_commit": "fb65467388c90a7d94d50fbcb711abfd8a568102", |
| "git_dirty": true |
| } |
| }, |
| "tokenizer_sha256": { |
| "tokenizer.json": "6155a4212832ddd656acd1cd45ab4acb327dbb080e3acdbec84f086adf17414e", |
| "tokenizer_config.json": "cccfc292e0846e8ab8aa04dafae32483a831bcf1df6f15100f82d306b4a9f4c6" |
| }, |
| "tensorboard_event_files": { |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869605.DESKTOP-OU3A33R.22736.0.eval": "09883b49cdd62a5c8d42822fc29a35820226c1a3dd208b40f9d979c36f869bea", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869629.DESKTOP-OU3A33R.22736.1.eval": "2e57b508c1a857eaa1ed999eed15938a410f7c121a210bad83fa69eeada93658", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869675.DESKTOP-OU3A33R.22736.2.eval": "2bae9a58f6c351803b0b319919a9d8cc7186ac3e2f2260ae30345a026a3e480c", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869703.DESKTOP-OU3A33R.22736.3.eval": "899e1194dc8114940e256510c7c05ae80ff677e26d37ae4bf993febf50dc7f34", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869755.DESKTOP-OU3A33R.22736.4.eval": "e6b697b446f79c28d37039be65f7b2c25ceb1229f2613fd0c14ae810bab0e69a", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790869765.DESKTOP-OU3A33R.22736.5.eval": "f1a8ba94bd9321bfe0e1c89de68577390172de30d014932a2cba5e418058bb5d", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790870472.DESKTOP-OU3A33R.22736.6.eval": "9fc3d6b6a242981d27e6de0c0ad4a6f3ebed56d14127b78b2d0d8ec4706378b3", |
| "runs/diffusion-gpu-v2/tb/events.out.tfevents.1790871766.DESKTOP-OU3A33R.22736.7.eval": "1d195bb7e90934ea6960c362e14f3d097376fa93db6bd3b141bf4b35f5eba525" |
| }, |
| "tensorboard_verified_points": 1166 |
| } |
| } |
|
|
