© 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. 16509
    Prev
    Next
    INDEX
    Explanations

    names that end with "vich" or "wicz"

    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
    anger
    -0.81
    rim
    -0.71
    angers
    -0.68
     Rugby
    -0.65
    mare
    -0.63
    Champ
    -0.63
    ANG
    -0.63
    ANGE
    -0.63
    acy
    -0.62
    Opt
    -0.62
    POSITIVE LOGITS
    vich
    3.50
    wic
    1.65
    л
    1.61
    zinski
    1.19
    vic
    1.12
    pta
    1.10
    witz
    1.10
     seated
    1.08
    cki
    1.03
    castle
    1.00
    Activations Density 0.061%

    No Known Activations