© 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. 25118
    Prev
    Next
    INDEX
    Explanations

    possessive forms and conversational phrases

    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
    478
    +0.17
    0.9%
    201
    +0.13
    0.7%
    23
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.17
    0.03
    134
    +0.13
    0.02
    433
    +0.13
    0.02
    Negative Logits
    ĻĤ
    -3.74
    £
    -3.68
    Ĵ
    -3.58
    ł
    -3.51
    ļ
    -3.49
    ĨĴ
    -3.42
    İ
    -3.39
    Ģ
    -3.30
    §
    -3.29
    ¢
    -3.29
    POSITIVE LOGITS
     gonna
    1.63
     gotta
    1.62
    dule
    1.48
     ital
    1.39
    istically
    1.37
     kidding
    1.37
     Exercise
    1.32
    ply
    1.31
     COPYRIGHT
    1.31
     yours
    1.29
    Activations Density 0.164%

    No Known Activations