© 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. 11-RES-JB
    11. 20745
    Prev
    Next
    INDEX
    Explanations

    numeric values or quantities in the context of rankings or levels of performance

    oai_token-act-pair · gpt-3.5-turbo

    references to prominent individuals and their achievements or characteristics

    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
    çīĪ
    -0.65
    ãĥ¯ãĥ³
    -0.65
    attribute
    -0.64
    ãĥĥ
    -0.63
     gist
    -0.61
     Cosponsors
    -0.61
    EStream
    -0.61
    chini
    -0.60
    ãĤ¦ãĤ¹
    -0.58
    200000
    -0.58
    POSITIVE LOGITS
     imaginable
    1.40
     alive
    1.28
     ever
    1.27
     EVER
    1.21
     anywhere
    1.12
     afloat
    0.96
     Alive
    0.96
    ever
    0.92
     around
    0.89
     conceivable
    0.87
    Activations Density 0.206%

    No Known Activations