© 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. 9591
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'W' in various formats and contexts

    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
    279
    +0.14
    0.8%
    114
    +0.13
    0.7%
    313
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    279
    +0.14
    0.05
    155
    +0.13
    0.03
    121
    +0.11
    0.03
    Negative Logits
    izumab
    -1.90
    urally
    -1.73
     totality
    -1.53
    unnumbered
    -1.45
     laundering
    -1.41
    ifically
    -1.40
     tertiary
    -1.38
    pective
    -1.38
     way
    -1.36
     schedule
    -1.35
    POSITIVE LOGITS
    instead
    1.99
    igg
    1.97
    inkles
    1.80
    inkle
    1.79
    Ĺ
    1.75
    irth
    1.73
    ohl
    1.69
    ¢
    1.64
    yler
    1.64
    ASHINGTON
    1.63
    Activations Density 0.099%

    No Known Activations