© 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. 24392
    Prev
    Next
    INDEX
    Explanations

    references to software and its licensing

    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
    87
    +0.12
    0.7%
    173
    +0.12
    0.7%
    71
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    506
    +0.12
    0.01
    74
    +0.12
    0.01
    490
    +0.10
    0.01
    Negative Logits
    áĢº
    -1.50
    ¯
    -1.47
    lor
    -1.43
    ³
    -1.42
    lining
    -1.42
    İ
    -1.34
    led
    -1.32
    áĢ¬
    -1.31
    iddell
    -1.31
    lying
    -1.31
    POSITIVE LOGITS
    beans
    1.97
     projects
    1.50
     appliances
    1.49
    stalk
    1.46
    plit
    1.44
     Bachelor
    1.43
    isan
    1.42
     inputs
    1.38
     alone
    1.38
    flower
    1.36
    Activations Density 0.048%

    No Known Activations