© 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. 16427
    Prev
    Next
    INDEX
    Explanations

    FAQs (Frequently Asked Questions) and related information

    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
    conn
    -0.92
    EStreamFrame
    -0.72
    Ò
    -0.68
    ivas
    -0.68
    Merit
    -0.67
    Root
    -0.67
    construct
    -0.67
     hashing
    -0.65
     infer
    -0.64
     milliseconds
    -0.64
    POSITIVE LOGITS
     FAQ
    2.50
     broch
    2.24
     pamphlet
    2.06
     booklet
    2.01
    FAQ
    1.75
     curriculum
    1.69
     Handbook
    1.69
     Guide
    1.68
     checklist
    1.65
     bulletin
    1.59
    Activations Density 0.038%

    No Known Activations