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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11-RES-JB
    6. 9652
    Prev
    Next
    INDEX
    Explanations

    sentences related to contracts, deals, and announcements

    oai_token-act-pair · gpt-3.5-turbo

    sentences that express significant conclusions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bias
    -0.68
     arsen
    -0.67
    iencies
    -0.65
     equivalents
    -0.63
     weighted
    -0.63
     probabilities
    -0.63
     polic
    -0.62
     fidelity
    -0.61
     transact
    -0.61
     civilians
    -0.60
    POSITIVE LOGITS
    Speaking
    1.13
    Born
    1.10
    Laun
    1.07
    Interested
    1.02
    Having
    0.99
    His
    0.97
    Following
    0.96
    After
    0.95
    He
    0.95
     Speaking
    0.94
    Activations Density 0.315%

    No Known Activations