© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12-RES-JB
    11. 2560
    Prev
    Next
    INDEX
    Explanations

    names and references to individuals, often in the context of serious situations

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    mentions of individuals involved in political discussions or events

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     pengu
    -0.94
     Milwaukee
    -0.85
     Guam
    -0.83
     Fal
    -0.82
     Baz
    -0.80
     Coco
    -0.80
     chefs
    -0.80
     Jupiter
    -0.77
     Gast
    -0.76
     Alb
    -0.74
    POSITIVE LOGITS
     Reid
    2.63
     Reeves
    1.44
     Peters
    1.19
     Nevada
    1.10
    Nev
    1.09
     Null
    1.04
     Levin
    1.03
     Pelosi
    0.98
     Ezra
    0.98
     Leh
    0.97
    Activations Density 0.278%

    No Known Activations