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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9451
    Prev
    Next
    INDEX
    Explanations

    numerical patterns in the format of "number-number"

    oai_token-act-pair · gpt-3.5-turbo

    numeric age references associated with individuals in a context that suggests their involvement in sports or related activities

    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
    cius
    -0.85
     sidewalks
    -0.77
     Thumbnails
    -0.76
     Hanna
    -0.73
    ollah
    -0.72
     FDR
    -0.71
     Zup
    -0.68
    imaru
    -0.68
     solicitation
    -0.67
     Olson
    -0.66
    POSITIVE LOGITS
    strong
    1.12
    match
    1.04
    year
    1.03
    league
    1.03
    goal
    1.00
    team
    0.99
    times
    0.98
    placed
    0.97
    month
    0.96
    player
    0.91
    Activations Density 0.063%

    No Known Activations