© 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. 26453
    Prev
    Next
    INDEX
    Explanations

    references to flavor or taste descriptors

    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
    406
    +0.15
    0.9%
    164
    +0.15
    0.8%
    369
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    406
    +0.15
    0.02
    164
    +0.15
    0.02
    354
    +0.12
    0.02
    Negative Logits
    chie
    -1.57
    ceans
    -1.56
    umerable
    -1.55
    ĸ´
    -1.55
    reon
    -1.51
    HECK
    -1.51
     knowing
    -1.49
    Ĥ¬
    -1.47
    »¿
    -1.46
     nowhere
    -1.44
    POSITIVE LOGITS
    ings
    1.86
    oxacin
    1.67
    attery
    1.54
    enders
    1.53
    awed
    1.53
    ameth
    1.50
    anked
    1.45
    authorized
    1.45
    avin
    1.43
    orence
    1.42
    Activations Density 0.038%

    No Known Activations