© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11-RES-JB
    11. 5043
    Prev
    Next
    INDEX
    Explanations

    instances of violent and criminal actions, particularly those involving assault, rape, stabbing, and attacking

    oai_token-act-pair · gpt-3.5-turbo

    terms related to sexual violence and assault

    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
     spons
    -0.79
    wcs
    -0.75
    WHERE
    -0.73
    iasm
    -0.71
    etheless
    -0.68
    profits
    -0.68
    taboola
    -0.66
    ãĥķãĤ©
    -0.63
    wcsstore
    -0.63
    suits
    -0.63
    POSITIVE LOGITS
     himself
    0.99
     him
    0.94
     someone
    0.92
     her
    0.89
     his
    0.88
     another
    0.87
     bystanders
    0.85
     classmate
    0.85
     somebody
    0.83
     Meredith
    0.83
    Activations Density 0.136%

    No Known Activations