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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 16522
    Prev
    Next
    INDEX
    Explanations

    the word "Stern"

    oai_token-act-pair · gpt-3.5-turbo
    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
     safely
    -0.80
    lane
    -0.74
    Cy
    -0.68
    dule
    -0.67
    ãģ«
    -0.67
    (*
    -0.66
     tree
    -0.62
     cycle
    -0.62
    æľ
    -0.61
     Afghanistan
    -0.61
    POSITIVE LOGITS
     Stern
    4.20
     Rosenberg
    1.17
     Bernstein
    1.12
     Eisen
    1.09
     Goldstein
    1.09
     Gross
    1.03
     Howard
    1.01
     Spiegel
    1.01
     Sloan
    1.00
     Weiner
    0.98
    Activations Density 0.019%

    No Known Activations