© 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. 27875
    Prev
    Next
    INDEX
    Explanations

    words related to squirming or squishing actions

    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
    369
    +0.18
    1.0%
    164
    +0.15
    0.9%
    338
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.18
    0.02
    338
    +0.15
    0.02
    164
    +0.15
    0.02
    Negative Logits
    inois
    -1.62
    quier
    -1.59
    poral
    -1.57
     fatal
    -1.56
    geant
    -1.56
    ational
    -1.53
    resa
    -1.50
     finding
    -1.47
    stant
    -1.46
    uginosa
    -1.45
    POSITIVE LOGITS
     (*
    1.65
    hill
    1.55
     (&
    1.48
    itzer
    1.41
    dog
    1.35
    fiddle
    1.30
    alled
    1.29
    awk
    1.25
    rant
    1.25
     [-
    1.24
    Activations Density 0.081%

    No Known Activations