© 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 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. 3-RES-JB
    11. 5731
    Prev
    Next
    INDEX
    Explanations

    instances of physical violence or injury to the face

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "face."

    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
    æ©Ł
    -0.76
    icultural
    -0.71
    icult
    -0.69
    ary
    -0.66
    imer
    -0.66
    itialized
    -0.65
    arity
    -0.65
    iculture
    -0.63
    CAST
    -0.63
    isol
    -0.62
    POSITIVE LOGITS
     face
    1.04
     faces
    0.98
    crow
    0.90
    plate
    0.89
    face
    0.87
    bish
    0.86
    plant
    0.81
    faces
    0.79
    plates
    0.77
     faced
    0.77
    Activations Density 0.029%

    No Known Activations