© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 10559
    Prev
    Next
    INDEX
    Explanations

    mentions of warranty in various contexts, particularly in legal or software licensing agreements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    78
    +0.17
    1.1%
    9
    +0.15
    0.9%
    397
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    495
    +0.17
    0.01
    52
    +0.15
    0.01
    216
    +0.12
    0.02
    Negative Logits
    ainen
    -1.78
    iser
    -1.48
    entric
    -1.38
    omo
    -1.38
    rapeut
    -1.36
    enta
    -1.30
    rapy
    -1.29
    ppler
    -1.28
    thon
    -1.25
    ctic
    -1.23
    POSITIVE LOGITS
     thereof
    1.98
    falls
    1.73
    fulness
    1.70
     whatsoever
    1.66
    аÑİ
    1.51
     implied
    1.40
    Ī
    1.35
     of
    1.34
     waiver
    1.32
    ī
    1.29
    Activations Density 0.070%

    No Known Activations