© 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. 1-RES-JB
    11. 9562
    Prev
    Next
    INDEX
    Explanations

    mentions of the word 'torture'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "tort" and its variations, indicating a focus on themes of torture or suffering

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     magnification
    -0.74
     Prospect
    -0.73
     Darkness
    -0.72
     Active
    -0.67
     Berlin
    -0.67
     Citizens
    -0.66
     brightest
    -0.63
     Czech
    -0.62
    FORE
    -0.62
     Copenhagen
    -0.61
    POSITIVE LOGITS
     tort
    1.33
    urous
    1.26
    oise
    1.18
    illas
    0.98
    uous
    0.94
    isec
    0.94
    ured
    0.94
    uring
    0.92
     Tort
    0.92
    onga
    0.89
    Activations Density 0.005%

    No Known Activations