{"bytes": 31535, "documents": 20, "environment": {"gpu": "NVIDIA GeForce RTX 3090", "platform": "Linux-7.0.0-34-generic-x86_64-with-glibc2.43", "python": "3.11.16 (main, Sep  1 2026, 14:18:37) [Clang 22.1.3 ]", "torch": "2.8.0+cu128", "transformers": "4.56.2"}, "finished": 1792198204, "gpu_slot_seconds": 1.042294049053453, "id": "3152ea52af35c64df2c4ac84", "model_load_seconds": 0.6576929540606216, "origin": "codesota", "parameters": 134515008, "peak_gpu_bytes": 1211815424, "per_document": [{"bytes": 1998, "id": "http://arxiv.org/abs/2610.22355v1", "nll_nats": 1326.2561619668268, "published": "2026-10-15T17:58:04Z", "sha256": "96d0c4ba95a814f10a3f81ebf67dc88f3bc96e2237bb66495dbb6f660352a609", "tokens": 345}, {"bytes": 1814, "id": "http://arxiv.org/abs/2610.22350v1", "nll_nats": 1395.6133055408282, "published": "2026-10-15T17:57:23Z", "sha256": "916c532294461331017aa54fb8a16a5b8c8759544c4d4bd1f7ecc3ef7cc647cd", "tokens": 369}, {"bytes": 1996, "id": "http://arxiv.org/abs/2610.22344v1", "nll_nats": 1606.6682546113443, "published": "2026-10-15T17:56:32Z", "sha256": "beac3e5550856e04e93cf053f7bf217da911bb628c57284ed725a5bfb231ff22", "tokens": 387}, {"bytes": 1322, "id": "http://arxiv.org/abs/2610.22333v1", "nll_nats": 980.5501083121053, "published": "2026-10-15T17:53:33Z", "sha256": "88189d65219571ca4301f4650c5456f11f94fb080523e3584d095955242d161c", "tokens": 265}, {"bytes": 1735, "id": "http://arxiv.org/abs/2610.22327v1", "nll_nats": 1251.849681863503, "published": "2026-10-15T17:52:18Z", "sha256": "24837e6678ef6dd34b06b89b681835c68b78b9818c9fafae5de6deb415892943", "tokens": 321}, {"bytes": 1529, "id": "http://arxiv.org/abs/2610.22324v1", "nll_nats": 1181.5504432769958, "published": "2026-10-15T17:52:12Z", "sha256": "035b9d4a48b9634d5e5024b7279e7fe869cb625151a5389ae5cd90d8e645271b", "tokens": 317}, {"bytes": 1687, "id": "http://arxiv.org/abs/2610.22322v1", "nll_nats": 1135.1231926808687, "published": "2026-10-15T17:51:57Z", "sha256": "a355314f1e2c5ac9e16a58d392d8d6206df7c1d76072ceb440804d16ee31d157", "tokens": 300}, {"bytes": 1675, "id": "http://arxiv.org/abs/2610.22314v1", "nll_nats": 1173.8500037786143, "published": "2026-10-15T17:51:01Z", "sha256": "aaef0080d8de7b12b1b38fa45b9be20ef78538715557f694575f8ad645691a46", "tokens": 307}, {"bytes": 1339, "id": "http://arxiv.org/abs/2610.22312v1", "nll_nats": 1025.0169193125112, "published": "2026-10-15T17:50:57Z", "sha256": "fe8b80c46a34756bcd97bdce924d18eebaccb264234a338c41051ba06cdd43b0", "tokens": 238}, {"bytes": 1973, "id": "http://arxiv.org/abs/2610.22300v1", "nll_nats": 1481.3895869016415, "published": "2026-10-15T17:47:52Z", "sha256": "735b57c61a0cc0f761ee3bb02d4a540d1bac6a57cc4da006e562f1bf0eaa8db7", "tokens": 388}, {"bytes": 1573, "id": "http://arxiv.org/abs/2610.22298v1", "nll_nats": 1395.0972015515435, "published": "2026-10-15T17:46:37Z", "sha256": "5f6b0ce3a295d552d75a4d40f4bd949d07be15bb5ea671fb6be7fdf0d272ac0e", "tokens": 308}, {"bytes": 1004, "id": "http://arxiv.org/abs/2610.22294v1", "nll_nats": 774.4290857721353, "published": "2026-10-15T17:44:46Z", "sha256": "e732845c2f51195f3f2c4d1b9c311c65c4dda091a0828d816689cbe95d7de915", "tokens": 180}, {"bytes": 1376, "id": "http://arxiv.org/abs/2610.22292v1", "nll_nats": 1137.015954777511, "published": "2026-10-15T17:43:47Z", "sha256": "9a2bfe5cc789c81b5302496c325e724f888c1960177661749ad564e5b313c29e", "tokens": 282}, {"bytes": 1599, "id": "http://arxiv.org/abs/2610.22273v1", "nll_nats": 1261.4414462978893, "published": "2026-10-15T17:35:17Z", "sha256": "728e16917f22536a6a8605a3adf5cc5abf5f0fb116da416d1d4ecd5d3dd60b97", "tokens": 296}, {"bytes": 1425, "id": "http://arxiv.org/abs/2610.22249v1", "nll_nats": 1256.2412406189833, "published": "2026-10-15T17:22:11Z", "sha256": "13ec28ad92281345c3c85675ca5e309ed96ee22a02020761bb7a90819ac2ea8f", "tokens": 291}, {"bytes": 1969, "id": "http://arxiv.org/abs/2610.22244v1", "nll_nats": 1927.7730661286187, "published": "2026-10-15T17:19:40Z", "sha256": "7a9b2b2878056194f0b652d7195a3738ad9f27e6be76cbdcd64893c9a9aab666", "tokens": 440}, {"bytes": 1339, "id": "http://arxiv.org/abs/2610.22236v1", "nll_nats": 1277.8378508752212, "published": "2026-10-15T17:15:01Z", "sha256": "f5b82a651f73956e080f02449036e630b990d3f15f42062e6cfa707a3e52e0ba", "tokens": 260}, {"bytes": 1617, "id": "http://arxiv.org/abs/2610.22209v1", "nll_nats": 1144.475001447514, "published": "2026-10-15T16:58:12Z", "sha256": "6103c7110886e56eece90f960ee60342947d1771b9f553b823e4088546870405", "tokens": 282}, {"bytes": 1306, "id": "http://arxiv.org/abs/2610.22196v1", "nll_nats": 1277.5483976423275, "published": "2026-10-15T16:46:47Z", "sha256": "9c4aba72a929ab63beb84949d16b39ffc5682c352dcc261ee11a998240cabf0c", "tokens": 286}, {"bytes": 1259, "id": "http://arxiv.org/abs/2610.22195v1", "nll_nats": 1013.2935713530751, "published": "2026-10-15T16:45:59Z", "sha256": "05c845b8dda718f80e1990434b2b0fe6d0fa0f9c55e23d0563466926fdf638a5", "tokens": 251}], "protocol": {"aggregation": "sum document negative log likelihood / log(2) / sum raw UTF-8 bytes", "batch_size": 4, "benchmark": "codesota-live-arxiv-2026-10-snapshot1", "context_tokens": 1024, "direction": "min", "dtype": "float32", "execution": "Native Slayer rows are unpadded and scored separately inside each batch; HF models use masked padded batches", "id": "codesota-live-bpb-v1", "loader_sources": {"sub1b_balanced.py": "f2949b3c2a6f3585ba9eb4b3bc7ef8306afea74492670d7c82d8c3f7e2c2a64b", "sub1b_gollem.py": "b9bff95fe326165b5685572727d5c50b1f86a50676932ad5b7da17018965e1d7", "sub1b_native.py": "ea56adc25566f7f6745ab882684dfc67e7f1cc48f29e53422fb3cab0cce769e7", "sub1b_slayer139.py": "f23a691bb330db4e6311b4b7e83936d5a4c584cc3f61ecf7f7025435ad87d326"}, "metric": "bits_per_utf8_byte", "runner_sha256": "1e4ab990f08eeffdce1a6202b97b8458913510106e96fcd0b22d37cffde37065", "selection": "20 newest cs.CL abstracts in pinned temporal manifest; domain-specific pilot, not representative general text", "start_token_id": 1, "start_token_policy": "HF tokenizer BOS, else EOS; native config BOS or token 0; prefix not scored", "stride_tokens": 512, "task": "language-modeling", "tf32": false, "tokenization": "raw text, no chat template or added special tokens; every content token scored once"}, "scope": "subset", "score": 1.144778422290382, "scoring_seconds": 0.38460109499283135, "scoring_tokens_per_second": 15894.390524587407, "system": {"adapter": "hf", "id": "HuggingFaceTB/SmolLM2-135M-Instruct", "license": "apache-2.0", "revision": "12fd25f77366fa6b3b4b768ec3050bf629380bac", "source": "https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct/tree/12fd25f77366fa6b3b4b768ec3050bf629380bac", "stage": "instruct", "track": "open"}, "tokens": 6113, "track": "open"}