© 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. 21183
    Prev
    Next
    INDEX
    Explanations

    references to various war crimes and atrocities

    oai_token-act-pair · gpt-3.5-turbo

    references to war crimes and atrocities

    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
     Slack
    -0.76
    ullivan
    -0.68
    agall
    -0.68
    alg
    -0.67
    iculty
    -0.66
    BuyableInstoreAndOnline
    -0.65
    Luck
    -0.65
    rame
    -0.64
     Chemistry
    -0.64
     Outlook
    -0.64
    POSITIVE LOGITS
     perpetrated
    1.68
     committed
    1.40
     atrocities
    1.30
     inflicted
    1.22
     genocide
    1.15
     massacres
    1.15
     crimes
    1.12
     impunity
    1.09
     against
    1.05
     perpetrators
    1.02
    Activations Density 0.267%

    No Known Activations