© 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. 31712
    Prev
    Next
    INDEX
    Explanations

    references to films, books, and the entertainment industry

    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
    156
    +0.18
    1.1%
    479
    +0.16
    0.9%
    321
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.18
    0.07
    469
    +0.16
    0.08
    340
    +0.15
    0.07
    Negative Logits
    ção
    -1.53
     quo
    -1.47
    otom
    -1.46
    rimination
    -1.41
    licts
    -1.39
    icable
    -1.36
    raints
    -1.36
     Clause
    -1.31
     veto
    -1.30
    ourced
    -1.28
    POSITIVE LOGITS
    ¥
    3.95
    ĻĤ
    3.62
    ij
    3.61
    ª
    3.56
    ¦
    3.51
    ¬
    3.44
    Ļ
    3.42
    ¨
    3.35
    ı
    3.34
    Ĺ
    3.34
    Activations Density 3.243%

    No Known Activations