© 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. 10193
    Prev
    Next
    INDEX
    Explanations

    words related to side effects in medical 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
    309
    +0.13
    0.8%
    337
    +0.12
    0.7%
    25
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    309
    +0.13
    0.03
    287
    +0.12
    0.03
    325
    +0.11
    0.03
    Negative Logits
    ient
    -1.94
    IST
    -1.55
    iente
    -1.51
    ging
    -1.46
     otherwise
    -1.44
    TING
    -1.42
    ty
    -1.42
    ienn
    -1.41
    itte
    -1.41
    ters
    -1.40
    POSITIVE LOGITS
    walks
    2.92
    bars
    2.55
    kick
    2.48
    walls
    2.35
    wall
    2.25
    bands
    2.13
    wise
    2.07
    arm
    1.99
    chains
    1.97
    chap
    1.81
    Activations Density 0.125%

    No Known Activations