© 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. 2-RES-SM
    11. 10613
    Prev
    Next
    INDEX
    Explanations

    words related to biochemical processes and relationships in health and disease

    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
    23
    +0.26
    1.5%
    321
    +0.17
    1.0%
    136
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.26
    0.08
    23
    +0.17
    0.08
    321
    +0.14
    0.06
    Negative Logits
     himself
    -1.91
    nament
    -1.74
     podium
    -1.67
     quarters
    -1.66
    ème
    -1.63
     sponsor
    -1.62
    buntu
    -1.61
    ista
    -1.61
     vendor
    -1.59
     card
    -1.56
    POSITIVE LOGITS
    Ī
    3.55
    »
    3.37
    ĸ
    3.35
    »¿
    3.28
    ¯
    3.27
    Ĺ
    3.24
    Į
    3.10
    ĥ½
    3.03
     ↵
    3.02
    3.02
    Activations Density 0.346%

    No Known Activations