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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    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. 23979
    Prev
    Next
    INDEX
    Explanations

    ages specified for activities or events targeted towards children or specific groups

    oai_token-act-pair · gpt-3.5-turbo

    terms and mentions related to age groups and demographics

    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
    saf
    -0.67
    Reloaded
    -0.65
    rious
    -0.63
    ilogy
    -0.62
    OLOG
    -0.62
     Massacre
    -0.59
     miracles
    -0.58
    forcement
    -0.58
     Rapp
    -0.57
    apego
    -0.57
    POSITIVE LOGITS
     alike
    1.11
     aged
    1.01
    pace
    1.00
    opausal
    0.97
     ages
    0.92
    paces
    0.91
    chool
    0.91
     wishing
    0.90
    hips
    0.88
     who
    0.87
    Activations Density 0.273%

    No Known Activations