© 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 EvalsExportsinterp-engineNEWAPI 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. 9889
    Prev
    Next
    INDEX
    Explanations

    reference to specific pages in a document

    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
    307
    +0.20
    1.1%
    241
    +0.12
    0.7%
    466
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    212
    +0.20
    0.02
    401
    +0.12
    0.01
    368
    +0.12
    0.01
    Negative Logits
    ¨
    -2.92
    ´
    -2.92
    §
    -2.80
    ĻĤ
    -2.69
    ¿
    -2.62
    ²
    -2.59
    Ĵ
    -2.56
    Īĺ
    -2.52
    ¼
    -2.35
    Ĥ¬
    -2.34
    POSITIVE LOGITS
    antry
    2.90
    views
    2.67
    view
    2.13
    render
    2.00
    fault
    1.96
    ant
    1.94
    less
    1.88
     reload
    1.85
    ants
    1.74
    page
    1.72
    Activations Density 0.209%

    No Known Activations