© 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. 9737
    Prev
    Next
    INDEX
    Explanations

    questions

    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
    307
    +0.42
    2.8%
    306
    +0.13
    0.9%
    258
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.42
    0.21
    307
    +0.13
    0.20
    400
    +0.11
    0.07
    Negative Logits
    bol
    -1.41
    boarding
    -1.36
    ligt
    -1.33
    ,^[@
    -1.32
    že
    -1.31
    '?"
    -1.31
    oretic
    -1.29
    ?"
    -1.28
    '?
    -1.26
    ?'
    -1.24
    POSITIVE LOGITS
     Lastly
    2.70
     Finally
    2.69
     Secondly
    2.21
     There
    2.17
     Also
    2.13
     Another
    2.08
     Conclusion
    2.05
     Then
    2.04
     Besides
    2.02
     Only
    2.02
    Activations Density 2.252%

    No Known Activations