© 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. 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. 11794
    Prev
    Next
    INDEX
    Explanations

    phrases related to refusal or rejection in official or formal contexts

    oai_token-act-pair · gpt-3.5-turbo

    cases of denial or refusal in various contexts

    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
     Travels
    -0.93
     Occupations
    -0.72
    issance
    -0.71
     intersect
    -0.70
     overflow
    -0.70
     invaluable
    -0.66
     ripple
    -0.65
     fascinating
    -0.63
    reetings
    -0.63
     Equip
    -0.62
    POSITIVE LOGITS
     preferring
    1.35
     fearing
    1.27
     citing
    1.18
     insisting
    1.07
     believing
    1.04
     vehemently
    1.03
     refused
    1.01
     opting
    1.01
     stating
    1.00
     discouraged
    0.95
    Activations Density 0.553%

    No Known Activations