© 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. 11324
    Prev
    Next
    INDEX
    Explanations

    syntax elements and data structure definitions typically found in programming code

    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
    21
    +0.43
    2.9%
    255
    +0.15
    1.0%
    450
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.43
    0.02
    21
    +0.15
    0.08
    185
    +0.11
    0.07
    Negative Logits
    -4.61
    <|outofrange|>
    -4.61
    č↵
    -4.61
    -4.61
     ↵
    -4.61
    -4.61
    -4.61
    ↵
    -4.61
    ↵  Âł
    -4.61
    ↵↵
    -4.61
    POSITIVE LOGITS
    (){
    1.60
    FromString
    1.55
     DEFAULT
    1.39
    Appellees
    1.38
    č↵č↵č
    1.37
    AtIndex
    1.34
    ById
    1.33
     Appellees
    1.32
     ///<
    1.31
     *)
    1.30
    Activations Density 1.440%

    No Known Activations