© 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. 24492
    Prev
    Next
    INDEX
    Explanations

    references to specific NFL teams and player positions

    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
    156
    +0.25
    1.4%
    81
    +0.15
    0.9%
    160
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.25
    0.04
    105
    +0.15
    0.04
    81
    +0.13
    0.04
    Negative Logits
    ĥ½
    -3.89
    ı
    -3.60
    ¦
    -3.47
    ¯
    -3.36
    ł
    -3.34
    Ĵ
    -3.31
    ±
    -3.30
    Ĥ
    -3.25
    º
    -3.24
    ¡
    -3.24
    POSITIVE LOGITS
    port
    2.37
    olve
    1.97
    pool
    1.95
    creen
    1.93
    uit
    1.93
    igan
    1.89
    istan
    1.89
    worth
    1.88
    apon
    1.84
    hell
    1.81
    Activations Density 0.255%

    No Known Activations