© 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. 11-RES-JB
    11. 5441
    Prev
    Next
    INDEX
    Explanations

    mentions of concrete actions or activities that are deemed problematic or dangerous

    oai_token-act-pair · gpt-3.5-turbo

    specific actions and behaviors related to health and safety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    achment
    -0.74
    ãĤ¬
    -0.71
    ãĤ·ãĥ£
    -0.68
    ija
    -0.67
    ortment
    -0.67
    agging
    -0.66
    dding
    -0.64
    ammy
    -0.64
    notations
    -0.63
    amation
    -0.63
    POSITIVE LOGITS
     themselves
    1.14
     oneself
    1.13
     infants
    0.94
     juveniles
    0.91
     unsuspecting
    0.91
     toddlers
    0.91
     teenagers
    0.90
     someone
    0.90
     strangers
    0.90
     inmates
    0.88
    Activations Density 0.344%

    No Known Activations