© 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. 12851
    Prev
    Next
    INDEX
    Explanations

    elements of code or programming syntax

    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
    177
    +0.15
    0.9%
    23
    +0.13
    0.7%
    136
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    136
    +0.15
    0.00
    177
    +0.13
    0.05
    21
    +0.10
    0.05
    Negative Logits
    arium
    -1.57
    áĢ¬
    -1.51
     programme
    -1.43
     consultation
    -1.41
    ashes
    -1.39
    zes
    -1.39
    documentclass
    -1.39
    versation
    -1.37
    omb
    -1.35
     lineup
    -1.34
    POSITIVE LOGITS
     ...,
    1.72
     respectively
    1.65
    acious
    1.42
     sort
    1.39
    ____________
    1.37
    liness
    1.36
     obviously
    1.35
    ldots
    1.33
     really
    1.32
     border
    1.32
    Activations Density 0.424%

    No Known Activations