© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 22481
    Prev
    Next
    INDEX
    Explanations

    references to witches and magical elements

    oai_token-act-pair · gpt-3.5-turbo

    references to witches and related themes

    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
     transcript
    -0.69
     é
    -0.69
    lez
    -0.68
    utations
    -0.67
    nikov
    -0.66
     hoc
    -0.65
    orters
    -0.64
     simultane
    -0.64
     seq
    -0.63
    reen
    -0.63
    POSITIVE LOGITS
     Witch
    3.35
     Goddess
    1.68
     Witches
    1.43
     Theme
    1.30
     Folk
    1.30
     Sorcerer
    1.27
     Psycho
    1.26
     Devil
    1.25
     Haunted
    1.24
     Wicked
    1.21
    Activations Density 0.048%

    No Known Activations