© 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. 27524
    Prev
    Next
    INDEX
    Explanations

    instances of initialization and installation instructions related to programming or software

    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
    203
    +0.13
    0.8%
    386
    +0.11
    0.7%
    309
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.02
    177
    +0.11
    0.02
    200
    +0.11
    0.02
    Negative Logits
    ĨĴ
    -2.04
    Ļª
    -2.03
    ĥ½
    -1.84
    ľĵ
    -1.72
    ®
    -1.69
    ypse
    -1.69
    £
    -1.68
    <|outofrange|>
    -1.67
    -1.67
    -1.67
    POSITIVE LOGITS
    ructions
    2.09
    ruction
    1.90
    arily
    1.89
    utions
    1.64
    alling
    1.60
    riction
    1.51
    agram
    1.51
     advanced
    1.46
    ulsion
    1.45
     up
    1.44
    Activations Density 0.235%

    No Known Activations