© 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. 2-RES-SM
    11. 13187
    Prev
    Next
    INDEX
    Explanations

    phrases related to biochemical processes and their effects on cells

    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
    47
    +0.21
    1.2%
    156
    +0.17
    1.0%
    240
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    47
    +0.21
    0.09
    275
    +0.17
    0.02
    11
    +0.15
    0.04
    Negative Logits
     banc
    -1.77
    giene
    -1.71
    arner
    -1.68
    ')->
    -1.67
    ertation
    -1.65
     profession
    -1.65
    apest
    -1.64
    riber
    -1.59
    pere
    -1.54
    ,—
    -1.54
    POSITIVE LOGITS
    Ĥ¬
    2.38
    ¦
    2.37
    Ń
    2.34
    ±
    2.33
    ĨĴ
    2.32
     endogenous
    2.32
    ¯
    2.30
    ¹
    2.23
    Ĩ
    2.23
    ł
    2.23
    Activations Density 1.104%

    No Known Activations