© 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. 0-RES-JB
    11. 11407
    Prev
    Next
    INDEX
    Explanations

    mentions of the word 'witch'

    oai_token-act-pair · gpt-3.5-turbo

    references to witches and witch hunts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Mobility
    -0.78
    Interstitial
    -0.77
     Transportation
    -0.73
     TAG
    -0.72
     Regions
    -0.70
     IMAGES
    -0.70
     Garage
    -0.69
    plates
    -0.68
     Bradley
    -0.67
    ERT
    -0.66
    POSITIVE LOGITS
     witch
    3.84
     Witch
    2.89
    witch
    2.62
     witches
    2.59
     Witches
    2.24
     witchcraft
    2.18
     sorce
    1.73
     sorcerer
    1.34
     Wizard
    1.34
     wizard
    1.33
    Activations Density 0.024%

    No Known Activations