© 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. 4-RES-SM
    11. 11129
    Prev
    Next
    INDEX
    Explanations

    code-related structures and keywords in 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/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
    25
    +0.25
    1.6%
    269
    +0.14
    0.9%
    308
    +0.12
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    308
    +0.25
    0.04
    25
    +0.14
    0.12
    330
    +0.12
    0.09
    Negative Logits
    ī
    -3.03
    Īĺ
    -2.88
    ĸ´
    -2.64
    ĨĴ
    -2.61
    į
    -2.61
    Ī
    -2.59
    ģ
    -2.52
    º
    -2.49
    Į
    -2.43
    FFIRMED
    -2.40
    POSITIVE LOGITS
    flies
    1.57
     journeys
    1.46
     journey
    1.44
     oneself
    1.41
     vehicles
    1.38
    overnment
    1.37
     vehicle
    1.35
     raft
    1.31
     drones
    1.31
    ipel
    1.28
    Activations Density 1.837%

    No Known Activations