© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 26349
    Prev
    Next
    INDEX
    Explanations

    instances of sharing and social interaction

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    17
    +0.20
    1.1%
    111
    +0.15
    0.8%
    377
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.20
    0.02
    57
    +0.15
    0.03
    197
    +0.12
    0.03
    Negative Logits
    ħ
    -2.40
    ¯
    -2.26
    Ī
    -2.21
    ©
    -1.97
    Ł
    -1.91
    ĥ
    -1.90
    ł
    -1.87
    ¨
    -1.85
    ĸ
    -1.81
    ģ
    -1.78
    POSITIVE LOGITS
    ibles
    1.90
     comments
    1.67
     Comments
    1.52
     Privacy
    1.50
     tips
    1.49
     chat
    1.49
    ible
    1.47
     Threat
    1.46
     Tribune
    1.46
     website
    1.44
    Activations Density 0.280%

    No Known Activations