© 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. 27357
    Prev
    Next
    INDEX
    Explanations

    instances of the word "reflect" and its variations

    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
    492
    +0.12
    0.6%
    304
    +0.11
    0.6%
    158
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.12
    0.01
    158
    +0.11
    0.02
    412
    +0.11
    0.03
    Negative Logits
    wered
    -1.45
    gg
    -1.43
    aan
    -1.42
    fs
    -1.42
    acht
    -1.35
     named
    -1.31
     sued
    -1.31
    arn
    -1.30
     file
    -1.28
    ustr
    -1.27
    POSITIVE LOGITS
    ulous
    1.73
     Areas
    1.57
    pectives
    1.56
    idian
    1.54
    adeon
    1.52
    rocal
    1.52
    ivity
    1.50
    IBILITY
    1.50
    IEW
    1.45
    onance
    1.45
    Activations Density 0.223%

    No Known Activations