© 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. 11584
    Prev
    Next
    INDEX
    Explanations

    words related to threats or warnings in a politically charged context

    oai_token-act-pair · gpt-3.5-turbo

    conditional statements involving threats or warnings

    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
    abal
    -0.77
    iosyncr
    -0.73
    arc
    -0.73
    ancer
    -0.68
    ochond
    -0.67
    azo
    -0.65
    ictionary
    -0.65
    WithNo
    -0.63
    Orig
    -0.63
    avascript
    -0.63
    POSITIVE LOGITS
    soever
    1.07
     elected
    0.92
     anyone
    0.83
     anybody
    0.77
     negotiators
    0.74
     they
    0.73
     convicted
    0.72
     policymakers
    0.72
     unsuccessful
    0.71
     necessary
    0.71
    Activations Density 0.112%

    No Known Activations