© 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 EvalsExportsAPI 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. 26074
    Prev
    Next
    INDEX
    Explanations

    instances of the word "previously."

    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
    404
    +0.11
    0.6%
    413
    +0.11
    0.6%
    397
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    206
    +0.11
    0.02
    15
    +0.11
    0.02
    404
    +0.11
    0.02
    Negative Logits
    ĨĴ
    -3.77
    ↵    ↵
    -3.68
    <|outofrange|>
    -3.68
    -3.68
    ↵↵
    -3.68
    ↵
    -3.68
       ↵
    -3.67
    -3.67
    ↵  âĢĥ
    -3.67
    -3.67
    POSITIVE LOGITS
    TRODUCTION
    1.68
     unpublished
    1.65
    born
    1.58
    occupied
    1.57
    noon
    1.54
    market
    1.53
     extinct
    1.53
    bra
    1.53
    grounds
    1.53
     published
    1.53
    Activations Density 0.026%

    No Known Activations