© 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. 3-RES-JB
    11. 17917
    Prev
    Next
    INDEX
    Explanations

    words related to illegal or immoral actions

    oai_token-act-pair · gpt-3.5-turbo

    references to crimes, particularly those described as severe or against humanity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    achev
    -0.72
    quickShipAvailable
    -0.68
    alg
    -0.68
    ihara
    -0.67
     Dak
    -0.65
    grad
    -0.65
    ulous
    -0.64
    arp
    -0.63
    whelming
    -0.61
    pole
    -0.61
    POSITIVE LOGITS
     committed
    0.93
     perpetrated
    0.91
     punishable
    0.89
     unfocusedRange
    0.88
     offences
    0.87
     spree
    0.86
     crimes
    0.84
     offenses
    0.84
    etting
    0.81
     abuses
    0.78
    Activations Density 0.019%

    No Known Activations