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. 3-RES-JB
    6. 19543
    Prev
    Next
    INDEX
    Explanations

    mentions of medical operations or procedures

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "minor" in various contexts and situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    romeda
    -0.77
    earchers
    -0.73
    adr
    -0.72
    Favorite
    -0.71
    ichick
    -0.69
    ardless
    -0.69
    oise
    -0.67
    reens
    -0.67
    ravings
    -0.65
    ++++++++++++++++
    -0.65
    POSITIVE LOGITS
    itized
    0.88
     accidents
    0.79
     leagues
    0.75
     inconven
    0.74
    ities
    0.74
     spoilers
    0.70
     minor
    0.70
     deviations
    0.69
     Subtle
    0.69
     inconvenience
    0.68
    Activations Density 0.011%

    No Known Activations