© 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. 0-RES-JB
    11. 11778
    Prev
    Next
    INDEX
    Explanations

    instances of a specific name, "Rasheed," and related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to authorities or formal statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     magnet
    -0.86
     cub
    -0.78
     canal
    -0.72
     populous
    -0.68
     Witcher
    -0.68
     fur
    -0.67
     lymph
    -0.67
     Cub
    -0.66
     sucked
    -0.66
     magnets
    -0.66
    POSITIVE LOGITS
    statement
    2.62
    heed
    2.12
    stat
    1.39
    stated
    1.36
    Statement
    1.26
     Lieberman
    1.18
    die
    1.17
    ISS
    1.09
     Statement
    1.09
    FG
    1.08
    Activations Density 0.042%

    No Known Activations