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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 0-RES-JB
    11. 11432
    Prev
    Next
    INDEX
    Explanations

    proper names or locations containing the substring "stad"

    oai_token-act-pair · gpt-3.5-turbo

    place names or locations associated with stadiums or sports arenas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     encrypted
    -0.68
     controlled
    -0.68
     stolen
    -0.66
     poison
    -0.65
     apps
    -0.63
     alloy
    -0.62
     relationship
    -0.61
     app
    -0.60
     programs
    -0.59
     workforce
    -0.59
    POSITIVE LOGITS
    stad
    5.02
    sted
    1.35
    lund
    1.25
    stadt
    1.24
    quist
    1.23
    ogle
    1.13
    holm
    1.11
    furt
    1.09
    stein
    1.07
    sson
    1.04
    Activations Density 0.030%

    No Known Activations