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. 1-RES-JB
    6. 9715
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically likely names or titles ending in 'her'

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific female individual or pronouns associated with her

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     reb
    -0.67
     Glou
    -0.65
     tie
    -0.62
     trim
    -0.61
     sur
    -0.61
     Rockefeller
    -0.60
     rolling
    -0.60
    ISO
    -0.59
     hypothetical
    -0.56
    RAW
    -0.56
    POSITIVE LOGITS
    itage
    1.44
    ald
    1.05
    theless
    1.03
    itance
    0.99
    ding
    0.94
    metic
    0.93
    rors
    0.88
    loo
    0.86
    rha
    0.86
    metics
    0.85
    Activations Density 0.010%

    No Known Activations