© 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. 12899
    Prev
    Next
    INDEX
    Explanations

    references to classes and their attributes within programming contexts

    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
    156
    +0.17
    0.9%
    2
    +0.16
    0.9%
    247
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    209
    +0.17
    0.04
    467
    +0.16
    0.03
    120
    +0.14
    0.03
    Negative Logits
    ĻĤ
    -4.18
    ª
    -3.95
    -3.94
    č↵
    -3.94
    ↵        ↵
    -3.94
    -3.94
    ↵↵
    -3.94
    -3.94
    <|outofrange|>
    -3.94
    -3.94
    POSITIVE LOGITS
    ifies
    2.23
    ifiers
    2.07
    ically
    1.79
    ifically
    1.72
    ify
    1.67
    ifying
    1.65
    ifier
    1.64
     itself
    1.63
     warfare
    1.59
    ifiable
    1.56
    Activations Density 0.094%

    No Known Activations