© 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. 11544
    Prev
    Next
    INDEX
    Explanations

    structured data or specific formatting characteristics in technical documents

    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
    324
    +0.14
    0.8%
    200
    +0.13
    0.7%
    78
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    315
    +0.14
    0.03
    324
    +0.13
    0.03
    200
    +0.11
    0.03
    Negative Logits
    hire
    -1.58
     practice
    -1.44
     Pradesh
    -1.43
     sense
    -1.41
     alone
    -1.39
    celand
    -1.38
     accompanies
    -1.38
     followed
    -1.37
     onward
    -1.34
    ipel
    -1.33
    POSITIVE LOGITS
    ī
    3.12
    ľĵ
    2.86
    ¥
    2.74
    Īĺ
    2.71
    į
    2.70
    ĨĴ
    2.66
    Ĩ
    2.66
    ĭ
    2.59
    Ķ
    2.58
    Ļª
    2.57
    Activations Density 0.143%

    No Known Activations