© 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. 1-RES-SM
    11. 27421
    Prev
    Next
    INDEX
    Explanations

    instances of the word "on."

    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
    69
    +0.21
    1.2%
    478
    +0.15
    0.9%
    452
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.21
    0.09
    243
    +0.15
    0.06
    36
    +0.13
    0.05
    Negative Logits
    ĸ
    -2.68
    Ĺ
    -2.49
    Ķ
    -2.49
    ľ
    -2.46
    Ī
    -2.45
    ¼
    -2.43
    Į
    -2.40
    ī
    -2.39
    ¬
    -2.31
    ²
    -2.31
    POSITIVE LOGITS
     behalf
    2.09
    shore
    2.09
    watson
    2.02
    amily
    1.94
    stage
    1.75
    dule
    1.73
    ahoma
    1.71
    cology
    1.69
    ego
    1.67
    yahoo
    1.66
    Activations Density 0.427%

    No Known Activations