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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9425
    Prev
    Next
    INDEX
    Explanations

    words related to criminal behavior and legal punishment

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals or groups that have been convicted of crimes

    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
    iasm
    -0.78
    ãĤ¡
    -0.77
     magnification
    -0.71
     mediation
    -0.71
     audible
    -0.70
    nergy
    -0.69
    mology
    -0.68
    erie
    -0.68
     Appro
    -0.66
     amplification
    -0.66
    POSITIVE LOGITS
     criminals
    1.52
     murderers
    1.50
     convicted
    1.48
     rapists
    1.44
     offenders
    1.40
     murderer
    1.36
     fug
    1.35
     rapist
    1.29
     fugitive
    1.24
     felon
    1.24
    Activations Density 0.314%

    No Known Activations