© 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. 24610
    Prev
    Next
    INDEX
    Explanations

    instances of the word "help" and its variations

    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
    341
    +0.13
    0.7%
    378
    +0.13
    0.7%
    365
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    378
    +0.13
    0.06
    341
    +0.13
    0.06
    237
    +0.12
    0.06
    Negative Logits
    pora
    -1.75
    áĢº
    -1.64
    ella
    -1.62
    elles
    -1.61
    etheless
    -1.57
    ço
    -1.55
    blic
    -1.53
     Islands
    -1.52
    finals
    -1.50
    ovirus
    -1.49
    POSITIVE LOGITS
    fully
    1.88
    º
    1.62
     us
    1.53
    ļ
    1.49
    ers
    1.48
    ĸ
    1.44
    ative
    1.42
    ward
    1.40
    anyl
    1.40
     clinicians
    1.39
    Activations Density 0.092%

    No Known Activations