© 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. 28825
    Prev
    Next
    INDEX
    Explanations

    individuals, particularly celebrities or significant persons

    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
    +0.24
    1.3%
    346
    +0.11
    0.6%
    50
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    369
    +0.24
    0.08
    457
    +0.11
    0.09
    346
    +0.10
    0.01
    Negative Logits
    other
    -1.91
    .^\[[@
    -1.81
    .),
    -1.78
    .^[@
    -1.71
    ipal
    -1.70
    others
    -1.68
    .\[[@
    -1.58
    .).
    -1.57
    selves
    -1.54
    .)
    -1.52
    POSITIVE LOGITS
    ↵
    3.38
       ↵
    3.38
    ↵
    3.38
    č↵
    3.38
    ↵Č
    3.38
    3.38
    č↵č↵
    3.38
    ↵  ³³³
    3.38
    <|outofrange|>
    3.38
    3.38
    Activations Density 2.091%

    No Known Activations