© 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. 4-RES-SM
    11. 14152
    Prev
    Next
    INDEX
    Explanations

    references to the "Star Wars" franchise

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    283
    +0.14
    0.8%
    275
    +0.12
    0.7%
    369
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.14
    0.03
    410
    +0.12
    0.03
    233
    +0.11
    0.02
    Negative Logits
     behalf
    -1.79
    uckland
    -1.67
    ullary
    -1.64
    mente
    -1.60
    usiness
    -1.56
    ément
    -1.54
    TRODUCTION
    -1.53
    ulates
    -1.50
     oxidase
    -1.50
    een
    -1.50
    POSITIVE LOGITS
     Trek
    3.04
    bucks
    2.92
    vation
    2.76
    burst
    2.68
     Wars
    2.56
    ved
    2.52
    ving
    2.46
    fle
    2.45
    light
    2.38
    buck
    2.19
    Activations Density 0.248%

    No Known Activations