© 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. 1-RES-SM
    11. 26617
    Prev
    Next
    INDEX
    Explanations

    references to the name "Dan."

    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
    369
    +0.25
    1.5%
    177
    +0.17
    1.0%
    211
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    211
    +0.25
    0.02
    369
    +0.17
    0.02
    177
    +0.14
    0.02
    Negative Logits
    >$
    -1.50
    ĥ½
    -1.47
    TRODUCTION
    -1.45
    rapeutic
    -1.44
    oretical
    -1.44
     drastic
    -1.40
    WISE
    -1.38
     different
    -1.36
    ODULE
    -1.35
     deaf
    -1.35
    POSITIVE LOGITS
    ube
    2.05
    nex
    1.92
    ico
    1.86
    ulo
    1.84
    iella
    1.79
    emann
    1.76
    iem
    1.73
    cer
    1.68
    ovich
    1.68
    zig
    1.68
    Activations Density 0.091%

    No Known Activations