© 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. 1-RES-JB
    11. 22515
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    flix
    -0.71
     adolesc
    -0.65
    clips
    -0.64
    bands
    -0.64
     Camel
    -0.64
     raids
    -0.63
    lude
    -0.63
    devices
    -0.63
    resses
    -0.63
     poisoning
    -0.62
    POSITIVE LOGITS
     Nanto
    0.76
    INGTON
    0.72
    ilitarian
    0.70
    milo
    0.67
    rolet
    0.66
     Originally
    0.66
    ãĥ¯ãĥ³
    0.64
     BI
    0.63
    ����
    0.62
    owski
    0.61
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.