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

    Neuronpedia

    Jacobian LensNEW
    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. 4-RES-JB
    11. 16949
    Prev
    Next
    INDEX
    Explanations

    mathematical values or percentages

    oai_token-act-pair · gpt-3.5-turbo

    quantifiable measures or significant numerical values

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

    No Comments

    Negative Logits
    vous
    -0.57
    Thumbnail
    -0.53
    uana
    -0.50
    itect
    -0.47
     Lesbian
    -0.47
     Sons
    -0.47
     Attribution
    -0.46
     Fathers
    -0.44
     Seat
    -0.44
     Defeat
    -0.44
    POSITIVE LOGITS
    orously
    0.63
    isively
    0.59
    apons
    0.57
    aneously
    0.56
    ishly
    0.55
    ensibly
    0.54
    onto
    0.53
    ibly
    0.52
    eatures
    0.52
    heartedly
    0.51
    Activations Density 1.599%

    No Known Activations