© 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. 1-RES-SM
    11. 27251
    Prev
    Next
    INDEX
    Explanations

    mentions of a specific person's name

    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
    241
    +0.16
    0.9%
    69
    +0.14
    0.8%
    144
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    241
    +0.16
    0.03
    101
    +0.14
    0.02
    321
    +0.13
    0.02
    Negative Logits
     anybody
    -1.65
     anyone
    -1.64
     );
    -1.61
    еÑģ
    -1.55
    =>
    -1.51
     .=
    -1.50
     anything
    -1.50
    ancy
    -1.47
    acles
    -1.47
    leftarrow
    -1.46
    POSITIVE LOGITS
    fter
    1.83
    heit
    1.79
    quez
    1.70
    xton
    1.69
    quel
    1.65
    bels
    1.65
    quet
    1.65
    uncher
    1.61
    pere
    1.59
    istry
    1.59
    Activations Density 0.147%

    No Known Activations