© 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. 10660
    Prev
    Next
    INDEX
    Explanations

    references to licenses

    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
    410
    +0.12
    0.7%
    244
    +0.12
    0.7%
    511
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.12
    0.01
    58
    +0.12
    0.01
    511
    +0.12
    0.01
    Negative Logits
    after
    -1.63
     fear
    -1.63
    stood
    -1.62
    closed
    -1.58
     order
    -1.58
    icity
    -1.51
    }</
    -1.50
    my
    -1.48
    other
    -1.47
    itro
    -1.46
    POSITIVE LOGITS
     Draft
    1.77
    etable
    1.75
    İ
    1.71
    lege
    1.69
    ¥
    1.67
     archives
    1.66
    ño
    1.65
     dissertation
    1.59
    Ľ
    1.58
    usiness
    1.57
    Activations Density 0.059%

    No Known Activations