# Language Modeling on FineWeb 10B (8K sequence length) Source: [Language Modeling on FineWeb 10B (8K sequence length)]() Collection index: [llms.txt]() Data updated: 2026-06-02T09:56:30.230Z Evaluation of language modeling performance on the FineWeb 10 Billion token dataset, constrained to an 8192 token sequence length, 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 8K 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 | | --- | --- | --- | --- | | LSSAR | 3.3171 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 3.8587 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 5.745 | Not recorded | 68fa7a8b1e209131cea83d15 | | LSSA | 5.9403 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N, l1-norm) | 6.1846 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 6.2339 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 6.2823 | Not recorded | 68fa7a8b1e209131cea83d15 | | LSSAR | 6.3007 | Not recorded | 68fa7a8b1e209131cea83d15 | | SSMax | 6.4384 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Li et al., 2022) | 6.4561 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (Tab. A.6) | 6.5701 | Not recorded | 68fa7a8b1e209131cea83d15 | | Softmax | 7.0183 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid (b = − log N) | 9.5939 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Wortsman et al., 2023) | 10.346 | Not recorded | 68fa7a8b1e209131cea83d15 | | ReLU (Shen et al., 2023) | 10.7266 | Not recorded | 68fa7a8b1e209131cea83d15 | | Sigmoid | 14.4995 | Not recorded | 68fa7a8b1e209131cea83d15 | ### LSSAR Validation Loss: 3.3171. Result ID: 6a1ea84afbc373592c19449f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N, l1-norm) Validation Loss: 3.8587. Result ID: 6a1ea84afbc373592c19449e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N, l1-norm) Validation Loss: 5.745. Result ID: 6a1ea84afbc373592c19449b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### LSSA Validation Loss: 5.9403. Result ID: 6a1ea84afbc373592c194499. [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: 6.1846. Result ID: 6a1ea84afbc373592c194494. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 6.2339. Result ID: 6a1ea84afbc373592c19449a. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 6.2823. Result ID: 6a1ea84afbc373592c194490. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### LSSAR Validation Loss: 6.3007. Result ID: 6a1ea84afbc373592c19449c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"p = 3"}. This record does not assert independent reproduction. ### SSMax Validation Loss: 6.4384. Result ID: 6a1ea84afbc373592c194491. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### ReLU (Li et al., 2022) Validation Loss: 6.4561. Result ID: 6a1ea84afbc373592c194497. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid (Tab. A.6) Validation Loss: 6.5701. Result ID: 6a1ea84afbc373592c194495. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Softmax Validation Loss: 7.0183. Result ID: 6a1ea84afbc373592c19449d. [Reporting paper]() Recorded evaluation configuration: {"Re-weighting":"p = 15"}. This record does not assert independent reproduction. ### Sigmoid (b = − log N) Validation Loss: 9.5939. Result ID: 6a1ea84afbc373592c194493. [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: 10.346. Result ID: 6a1ea84afbc373592c194496. [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: 10.7266. Result ID: 6a1ea84afbc373592c194498. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. ### Sigmoid Validation Loss: 14.4995. Result ID: 6a1ea84afbc373592c194492. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Re-weighting":"none"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.