© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 22041
    Prev
    Next
    INDEX
    Explanations

    names of people and organizations, particularly focusing on specific positions or titles

    oai_token-act-pair · gpt-3.5-turbo

    instances of negative or harmful situations or behaviors

    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
     NEXT
    -0.81
     FIGHT
    -0.78
     sustaining
    -0.77
     rooting
    -0.76
     conflicts
    -0.76
     acronym
    -0.75
     consist
    -0.74
     shaping
    -0.71
     conflict
    -0.71
     exper
    -0.70
    POSITIVE LOGITS
    Smith
    1.72
    Jones
    1.69
    Williams
    1.60
    Brown
    1.59
    Hill
    1.57
    Harris
    1.55
    Thompson
    1.54
    Roberts
    1.54
    Johnson
    1.52
    Davis
    1.50
    Activations Density 0.038%

    No Known Activations