© 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. 23863
    Prev
    Next
    INDEX
    Explanations

    actions involving legal or illegal activities

    oai_token-act-pair · gpt-3.5-turbo

    actions related to illegal activities or misconduct

    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
    oÄŁ
    -0.73
     wondering
    -0.67
    sson
    -0.67
     reckoning
    -0.65
    witz
    -0.64
    inav
    -0.64
    depending
    -0.62
    redo
    -0.62
    endif
    -0.61
    enment
    -0.61
    POSITIVE LOGITS
     indecent
    1.18
     pornographic
    1.10
     obscene
    1.09
     improper
    1.05
     prostitutes
    1.00
     illegal
    0.99
     inappropriate
    0.98
     unauthorized
    0.97
     bribes
    0.96
     unlawful
    0.96
    Activations Density 0.252%

    No Known Activations