© 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. 2-RES-SM
    11. 13574
    Prev
    Next
    INDEX
    Explanations

    references to legal citations and case law

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    221
    +0.11
    0.6%
    222
    +0.10
    0.6%
    374
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    374
    +0.11
    0.04
    175
    +0.10
    0.03
    221
    +0.09
    0.03
    Negative Logits
    ?”
    -1.58
    ?’
    -1.53
    $.[]{
    -1.47
    ...](
    -1.44
     rescue
    -1.43
    break
    -1.33
     drill
    -1.33
    !”
    -1.31
    ?'
    -1.31
     goal
    -1.31
    POSITIVE LOGITS
    Ŀ
    1.81
    bered
    1.74
    idable
    1.73
    witz
    1.66
    ¦
    1.61
    oyer
    1.60
    dorff
    1.50
    taire
    1.47
    ¯
    1.47
    ões
    1.45
    Activations Density 0.093%

    No Known Activations