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. 7328
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific topics or themes, such as "Islam and the Founders," "alcohol and health outcomes," and "women and the economy."

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize connection and relationships

    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
    DERR
    -0.75
    burgh
    -0.71
    Lt
    -0.70
    ainer
    -0.70
    avorite
    -0.69
    Cmd
    -0.68
    umo
    -0.68
    Alert
    -0.67
    DER
    -0.67
    atel
    -0.66
    POSITIVE LOGITS
     spirituality
    0.94
     politics
    0.93
     morality
    0.91
     sexuality
    0.89
     destiny
    0.85
     culture
    0.85
     Politics
    0.85
     psychology
    0.83
     manners
    0.82
     masculinity
    0.82
    Activations Density 0.250%

    No Known Activations