© 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. 24671
    Prev
    Next
    INDEX
    Explanations

    questions and rhetorical devices

    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
    23
    +0.31
    1.8%
    410
    +0.14
    0.8%
    365
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.31
    0.13
    23
    +0.14
    0.13
    369
    +0.10
    0.08
    Negative Logits
    ?,
    -1.78
    ?",
    -1.76
    )?
    -1.69
    )",
    -1.65
    liga
    -1.65
    ernel
    -1.61
     reprodu
    -1.59
    ?)
    -1.57
    amento
    -1.57
     cult
    -1.56
    POSITIVE LOGITS
    ·
    2.90
    ¿½
    2.88
     ↵
    2.85
    2.85
    2.85
    ↵↵
    2.85
    2.85
    ↵
    2.85
    2.85
    2.85
    Activations Density 1.458%

    No Known Activations