© 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. 1-RES-SM
    11. 24273
    Prev
    Next
    INDEX
    Explanations

    references to shaders and related graphics programming concepts

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    +0.23
    1.4%
    144
    +0.17
    1.0%
    100
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    144
    +0.23
    0.02
    100
    +0.17
    0.01
    43
    +0.13
    0.01
    Negative Logits
    woke
    -1.91
    ffer
    -1.70
    vas
    -1.68
     great
    -1.54
    ften
    -1.52
    UGH
    -1.46
     somebody
    -1.45
    redit
    -1.45
    tox
    -1.43
     dass
    -1.43
    POSITIVE LOGITS
    ière
    1.86
    enberg
    1.75
    ager
    1.74
    sburg
    1.71
    smen
    1.69
     altogether
    1.69
    naire
    1.68
    works
    1.67
     thereof
    1.64
    ing
    1.63
    Activations Density 0.093%

    No Known Activations