© 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. 28918
    Prev
    Next
    INDEX
    Explanations

    instances of the word "analyze" and its variations, indicating a focus on analysis and assessment

    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
    23
    +0.20
    1.1%
    156
    +0.19
    1.1%
    21
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.20
    0.01
    345
    +0.19
    0.01
    412
    +0.14
    0.01
    Negative Logits
    ŀ
    -2.15
    ĸ
    -2.11
    als
    -1.88
    walks
    -1.69
    §
    -1.69
    Ľ
    -1.67
    ĭ
    -1.66
    IJ
    -1.62
    enberg
    -1.59
    ®
    -1.59
    POSITIVE LOGITS
    estion
    1.77
    apore
    1.51
     him
    1.50
     retrospectively
    1.49
     seeing
    1.49
     sight
    1.49
    them
    1.43
     quantitatively
    1.42
     them
    1.42
    just
    1.42
    Activations Density 0.046%

    No Known Activations