# Language Modeling on FineWeb 10B (1K sequence length, val) Source: [Language Modeling on FineWeb 10B (1K sequence length, val)]() Collection index: [llms.txt]() Data updated: 2026-06-02T09:56:30.120Z Evaluation of language model performance on the FineWeb 10 Billion token dataset, specifically restricted to sequences of 1,000 tokens, measured by Validation Loss. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Language Modeling | | Dataset (including any stated split) | FineWeb 10B 1K sequence length | | Metric | Validation Loss | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Validation Loss](): Lower is better; unit: SCALAR. ## Recorded results | Method | Validation Loss | Result date | Report paper | | --- | --- | --- | --- | | SSMax | 3.1705 | Not recorded | 68fa7a8b1e209131cea83d15 | | LSSAR | 3.1782 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 3.1841 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 3.1849 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 3.1879 | Not recorded | 68fa7a8b1e209131cea83d15 | | LSSA | 3.1905 | Not recorded | 68fa7a8b1e209131cea83d15 | | LSSAR | 3.1905 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 3.1911 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N) | 3.193 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid | 3.1935 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (Tab. A.6) | 3.2 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Li et al., 2022) | 3.2006 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Wortsman et al., 2023) | 3.2143 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Shen et al., 2023) | 3.2155 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 3.2171 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 5.3878 | Not recorded | 68fa7a8b1e209131cea83d15 | ### SSMax Validation Loss: 3.1705. Result ID: 6a1ea84afbc373592c194461. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### LSSAR Validation Loss: 3.1782. Result ID: 6a1ea84afbc373592c19446c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N, l1-norm) Validation Loss: 3.1841. Result ID: 6a1ea84afbc373592c19446b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N, l1-norm) Validation Loss: 3.1849. Result ID: 6a1ea84afbc373592c194464. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 3.1879. Result ID: 6a1ea84afbc373592c19446a. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### LSSA Validation Loss: 3.1905. Result ID: 6a1ea84afbc373592c194469. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### LSSAR Validation Loss: 3.1905. Result ID: 6a1ea84afbc373592c19446f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 3.1911. Result ID: 6a1ea84afbc373592c194460. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N) Validation Loss: 3.193. Result ID: 6a1ea84afbc373592c194463. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid Validation Loss: 3.1935. Result ID: 6a1ea84afbc373592c194462. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid (Tab. A.6) Validation Loss: 3.2. Result ID: 6a1ea84afbc373592c194465. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### ReLU (Li et al., 2022) Validation Loss: 3.2006. Result ID: 6a1ea84afbc373592c194467. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### ReLU (Wortsman et al., 2023) Validation Loss: 3.2143. Result ID: 6a1ea84afbc373592c194466. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### ReLU (Shen et al., 2023) Validation Loss: 3.2155. Result ID: 6a1ea84afbc373592c194468. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N, l1-norm) Validation Loss: 3.2171. Result ID: 6a1ea84afbc373592c19446e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 5.3878. Result ID: 6a1ea84afbc373592c19446d. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.