Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 19974
    Prev
    Next
    INDEX
    Explanations

    references to sports leagues and players

    oai_token-act-pair · gpt-3.5-turbo

    statements about events or outcomes related to sports and injuries

    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
     Spir
    -0.94
     Qué
    -0.87
     typh
    -0.87
    £ı
    -0.83
     Jes
    -0.83
     raft
    -0.81
    Äį
    -0.81
    jac
    -0.80
     republican
    -0.77
     Monteneg
    -0.77
    POSITIVE LOGITS
     Allen
    2.53
    Allen
    2.47
     Allan
    1.24
     Kirk
    1.22
     Wade
    1.07
     Ald
    1.04
     Andy
    1.03
     Bryant
    1.01
     Orion
    1.00
     Woody
    0.98
    Activations Density 0.189%

    No Known Activations