© 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 EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 11450
    Prev
    Next
    INDEX
    Explanations

    locations and facilities related to confinement or storage

    oai_token-act-pair · gpt-3.5-turbo

    terms related to observation and rescue operations in the context of detention or emergency situations

    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
    Ô
    -0.70
    ethe
    -0.68
    etically
    -0.66
    acus
    -0.64
     Curry
    -0.63
    ounced
    -0.62
    uate
    -0.62
    oiler
    -0.62
    ãĥ£
    -0.61
    xxxxxxxx
    -0.61
    POSITIVE LOGITS
     station
    1.14
     stations
    1.08
    station
    1.00
     facility
    0.99
     centres
    0.98
     Facility
    0.95
     centers
    0.94
     depot
    0.94
     bureau
    0.93
     facilities
    0.93
    Activations Density 0.324%

    No Known Activations