© 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. 2-RES-SM
    11. 29785
    Prev
    Next
    INDEX
    Explanations

    variations of the word "make" and its various forms or uses

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    397
    +0.16
    0.9%
    407
    +0.14
    0.8%
    317
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    397
    +0.16
    0.07
    303
    +0.14
    0.07
    317
    +0.12
    0.06
    Negative Logits
    ij
    -2.60
    Ĥ¬
    -2.51
    ¬
    -2.26
    ĨĴ
    -2.25
    ı
    -2.10
    Ń
    -2.10
    ī
    -2.07
    Ľ
    -2.06
    ¿
    -2.02
    İ
    -1.96
    POSITIVE LOGITS
    hift
    2.50
    leine
    2.29
     haste
    1.78
     merry
    1.50
    pless
    1.43
    ercase
    1.40
    ERN
    1.40
    urate
    1.39
    per
    1.36
    eligible
    1.35
    Activations Density 0.168%

    No Known Activations