© 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. 4-RES-SM
    11. 11051
    Prev
    Next
    INDEX
    Explanations

    expressions related to physical activity or sports

    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
    397
    +0.15
    0.8%
    263
    +0.14
    0.8%
    272
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    263
    +0.15
    0.20
    56
    +0.14
    0.07
    117
    +0.13
    0.15
    Negative Logits
    ĩ
    -7.02
    ·
    -6.83
    Ļª
    -6.78
    ↵
    -6.75
    ↵
    -6.75
           ↵
    -6.75
    ↵
    -6.75
    č↵
    -6.75
    -6.75
    č↵
    -6.75
    POSITIVE LOGITS
     myself
    3.38
     my
    3.10
     lol
    2.67
     :(
    2.53
     me
    2.47
     though
    2.43
     LOL
    2.32
     :)
    2.31
     :-)
    2.29
     I
    2.28
    Activations Density 3.628%

    No Known Activations