© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 12334
    Prev
    Next
    INDEX
    Explanations

    references to legal or political documents and specific terminology associated with formal communication

    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
    198
    +0.14
    0.8%
    207
    +0.12
    0.7%
    34
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    330
    +0.14
    0.07
    198
    +0.12
    0.07
    68
    +0.11
    0.07
    Negative Logits
     authors
    -1.91
    '];
    -1.85
    ...](
    -1.74
     missing
    -1.58
    >(
    -1.56
    >();
    -1.53
     author
    -1.51
     availability
    -1.49
     prey
    -1.46
     courage
    -1.44
    POSITIVE LOGITS
    wikipedia
    2.02
    eren
    1.71
    mere
    1.69
     arsenal
    1.68
     context
    1.61
    bilt
    1.57
    ague
    1.52
    hold
    1.46
     Hills
    1.45
     regard
    1.44
    Activations Density 1.347%

    No Known Activations