© 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. 13923
    Prev
    Next
    INDEX
    Explanations

    the word "just" in various 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
    201
    +0.12
    0.7%
    378
    +0.12
    0.7%
    140
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    110
    +0.12
    0.07
    140
    +0.12
    0.06
    201
    +0.11
    0.06
    Negative Logits
    ĨĴ
    -1.85
    İ
    -1.74
    holder
    -1.69
    ĭ
    -1.68
    Ļª
    -1.65
    Ĵ
    -1.61
    soever
    -1.56
    atics
    -1.52
    msgstr
    -1.50
    ĸ´
    -1.50
    POSITIVE LOGITS
    ifiable
    2.96
    ifications
    2.69
    ifi
    2.69
    ifying
    2.37
    ifies
    2.16
    icia
    2.13
    ification
    2.06
    ified
    1.96
    iciary
    1.96
    icar
    1.85
    Activations Density 0.178%

    No Known Activations