© 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. 11863
    Prev
    Next
    INDEX
    Explanations

    references to criminal activities and legal terms

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with legal convictions and 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
    izons
    -0.81
     spotlight
    -0.75
    ellen
    -0.69
     reinforcements
    -0.65
     ears
    -0.64
     partName
    -0.64
    oscopic
    -0.64
    enhagen
    -0.63
     narrator
    -0.63
    Tokens
    -0.63
    POSITIVE LOGITS
     crimes
    1.92
     offences
    1.80
     offenses
    1.72
     misdem
    1.72
     crime
    1.70
    crime
    1.54
     heinous
    1.54
     offence
    1.50
     Crimes
    1.50
     wrongdoing
    1.50
    Activations Density 0.375%

    No Known Activations