© 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. 10094
    Prev
    Next
    INDEX
    Explanations

    instances of the word "find" and related phrases indicating the search for information or solutions

    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
    +0.27
    1.5%
    303
    +0.10
    0.6%
    52
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    230
    +0.27
    0.02
    339
    +0.10
    0.02
    307
    +0.10
    0.05
    Negative Logits
    isters
    -1.48
    ultz
    -1.45
    hip
    -1.42
    NESS
    -1.40
    oat
    -1.29
     ambitions
    -1.28
    ymoon
    -1.26
     invade
    -1.26
    quart
    -1.26
    opolys
    -1.24
    POSITIVE LOGITS
     answers
    2.37
     anywhere
    2.36
     answer
    2.20
     solutions
    2.09
     solution
    2.04
    Ļ
    1.96
     suitable
    1.95
     way
    1.94
     examples
    1.93
     tutorials
    1.87
    Activations Density 0.383%

    No Known Activations