© 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. 9437
    Prev
    Next
    INDEX
    Explanations

    phrases associated with gaming and competitiveness

    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
    156
    +0.37
    2.2%
    307
    +0.20
    1.2%
    271
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    181
    +0.37
    0.09
    386
    +0.20
    0.02
    146
    +0.13
    0.09
    Negative Logits
     """
    -1.54
     Corinth
    -1.41
     ".
    -1.39
     _________
    -1.33
    __.
    -1.29
     polyg
    -1.29
     _____
    -1.28
     BACKGROUND
    -1.28
    ↵
    -1.28
           ↵
    -1.28
    POSITIVE LOGITS
    leneck
    2.13
    irubin
    1.79
    acker
    1.79
    ename
    1.78
    imoto
    1.78
    icode
    1.76
    ciparum
    1.72
    äsident
    1.68
    ings
    1.67
    aga
    1.67
    Activations Density 0.743%

    No Known Activations