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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4-RES-JB
    11. 15374
    Prev
    Next
    INDEX
    Explanations

    references to talk show hosts, especially Jimmy Kimmel and Stephen Colbert

    oai_token-act-pair · gpt-3.5-turbo

    mentions of late-night talk show hosts, particularly Stephen Colbert and Jimmy Kimmel

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

    No Comments

    Negative Logits
    ramid
    -0.73
    ntil
    -0.72
    arching
    -0.70
    arma
    -0.70
    axter
    -0.70
    esthetic
    -0.69
    board
    -0.69
    Reviewer
    -0.69
    boarding
    -0.67
     Finnish
    -0.67
    POSITIVE LOGITS
     Colbert
    1.07
    vill
    0.80
     Fallon
    0.76
    opard
    0.76
     Comedy
    0.71
     Presents
    0.70
    wald
    0.69
     memes
    0.68
     Stewart
    0.68
    olini
    0.67
    Activations Density 0.010%

    No Known Activations