Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 23232
    Prev
    Next
    INDEX
    Explanations

    the adjective "bold" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "bold" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    OTOS
    -0.92
     Cheong
    -0.88
    yip
    -0.68
    enfranch
    -0.64
    utra
    -0.63
     duly
    -0.62
    rogens
    -0.62
    RT
    -0.61
    ADS
    -0.61
    nesota
    -0.60
    POSITIVE LOGITS
    faced
    1.21
    er
    1.09
    face
    1.04
    ness
    1.04
    mouth
    0.90
    bold
    0.87
     bold
    0.86
    est
    0.86
    word
    0.85
    words
    0.84
    Activations Density 0.032%

    No Known Activations