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. 10-RES-JB
    6. 15645
    Prev
    Next
    INDEX
    Explanations

    mentions of women and sometimes children in various contexts, including scenarios of assault, labor supply, and community roles

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Reviewer
    -0.82
    ourney
    -0.73
     guiActiveUnfocused
    -0.68
     Prol
    -0.68
    mpire
    -0.67
    went
    -0.65
     Scare
    -0.65
     Hoover
    -0.64
     Collider
    -0.64
     Logged
    -0.64
    POSITIVE LOGITS
     minorities
    1.48
     girls
    1.37
     children
    1.32
     infants
    1.19
     gays
    1.14
     lesbians
    1.14
     women
    1.13
     Girls
    1.12
     babies
    1.11
     adolescents
    1.11
    Activations Density 0.103%

    No Known Activations