© 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. 1-RES-SM
    11. 28393
    Prev
    Next
    INDEX
    Explanations

    negations and phrases indicating absence or lack

    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
    83
    +0.14
    0.7%
    313
    +0.11
    0.6%
    422
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    83
    +0.14
    0.04
    313
    +0.11
    0.03
    9
    +0.11
    0.02
    Negative Logits
    ±
    -2.23
    ĭ
    -2.22
    ½
    -2.15
    ©
    -2.08
    ģ
    -2.08
    ¾
    -2.06
    ¹
    -2.04
    imum
    -2.04
    Ī
    -1.99
    Ĵ
    -1.97
    POSITIVE LOGITS
    coming
    1.53
    here
    1.43
     Utah
    1.38
     Saints
    1.38
     Lions
    1.38
    born
    1.35
     virgin
    1.35
    abol
    1.35
     meantime
    1.34
     doubts
    1.31
    Activations Density 0.455%

    No Known Activations