© 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. 8-RES-JB
    11. 19303
    Prev
    Next
    INDEX
    Explanations

    instances where the phrase "stick" (or variations like "sticking") is used in a document

    oai_token-act-pair · gpt-3.5-turbo

    the word "stick" in various contexts related to adherence or continuity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unfocusedRange
    -0.73
    DragonMagazine
    -0.70
    æ©Ł
    -0.66
     Pt
    -0.62
    bered
    -0.61
    Planet
    -0.61
    çļĦ
    -0.60
    une
    -0.60
    Export
    -0.59
    KEN
    -0.58
    POSITIVE LOGITS
     plaster
    0.88
    river
    0.82
    rily
    0.80
    handle
    0.78
     blender
    0.77
    tails
    0.75
    holder
    0.74
    holders
    0.73
     stick
    0.71
     toe
    0.71
    Activations Density 0.021%

    No Known Activations