© 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. 1-RES-SM
    11. 25227
    Prev
    Next
    INDEX
    Explanations

    references to a specific person or character named "Parnell."

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    118
    +0.15
    0.8%
    369
    +0.11
    0.6%
    410
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    118
    +0.15
    0.02
    146
    +0.11
    0.01
    40
    +0.11
    0.02
    Negative Logits
    olation
    -1.65
    OL
    -1.62
    ulmonary
    -1.52
    WD
    -1.50
    rology
    -1.47
    WHO
    -1.44
    PSC
    -1.43
    gree
    -1.43
    meric
    -1.42
    rench
    -1.41
    POSITIVE LOGITS
    İ
    3.22
    ŀ
    3.18
    ©
    3.09
    Ł
    3.08
    ¢
    2.96
    ¨
    2.86
    Ĵ
    2.85
    ł
    2.83
    ĩ
    2.80
    ↵
    2.79
    Activations Density 0.090%

    No Known Activations