© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 1066
    Prev
    Next
    INDEX
    Explanations

    open-ended questions starting with "who knows" or related variations

    oai_token-act-pair · gpt-3.5-turbo

    expressions of uncertainty and curiosity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     moot
    -0.70
     guiActiveUnfocused
    -0.66
    aneous
    -0.65
     induction
    -0.63
     cancellation
    -0.63
    naissance
    -0.62
    arily
    -0.61
    ooters
    -0.61
    vity
    -0.60
    reen
    -0.59
    POSITIVE LOGITS
    ?),
    0.79
     hect
    0.72
    ?)
    0.72
    ij士
    0.68
     Cameroon
    0.68
    opsy
    0.66
    aca
    0.65
     Jagu
    0.63
    uld
    0.62
    jah
    0.62
    Activations Density 0.191%

    No Known Activations