© 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. 2-RES-SM
    11. 9735
    Prev
    Next
    INDEX
    Explanations

    commands and functions related to shell scripting

    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
    369
    +0.16
    0.9%
    334
    +0.14
    0.8%
    478
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    334
    +0.16
    0.03
    215
    +0.14
    0.03
    131
    +0.12
    0.02
    Negative Logits
    hire
    -1.64
    ]{}.
    -1.59
     conceived
    -1.50
    *](#
    -1.49
    *]{}.
    -1.47
    )\].
    -1.45
    ptin
    -1.44
    )\]
    -1.44
    ordan
    -1.43
    tera
    -1.41
    POSITIVE LOGITS
     warnings
    1.78
     '<
    1.59
     hello
    1.57
     positivity
    1.55
     "[
    1.51
     plain
    1.50
     "\
    1.48
     Invalid
    1.47
     preponderance
    1.46
     '[
    1.45
    Activations Density 0.184%

    No Known Activations