© 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. 1-RES-JB
    11. 11184
    Prev
    Next
    INDEX
    Explanations

    references to specific entities or concepts, such as "Jedi" and "Sith"

    oai_token-act-pair · gpt-3.5-turbo

    references to the "Vueltas" and "Jedi" in contexts related to racing and Star Wars

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tracks
    -0.85
    ball
    -0.77
    myra
    -0.73
    sheet
    -0.73
    iday
    -0.71
    lot
    -0.71
    sy
    -0.70
    bath
    -0.69
    tip
    -0.69
    erences
    -0.69
    POSITIVE LOGITS
     Vu
    0.89
     Lumpur
    0.76
    ulsion
    0.75
    Lens
    0.75
     dism
    0.71
    ¥ŀ
    0.70
     Sabha
    0.70
    itton
    0.70
     conflic
    0.69
    à¥
    0.67
    Activations Density 0.013%

    No Known Activations