© 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. 0-RES-JB
    11. 9704
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "raz" or words containing "raz" with varying intensities

    oai_token-act-pair · gpt-3.5-turbo

    the presence of specific names or terms, particularly those related to individuals or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     aff
    -0.68
     grammar
    -0.67
     uncond
    -0.65
     FG
    -0.65
     English
    -0.62
     Edinburgh
    -0.61
     ES
    -0.61
     appearance
    -0.61
     organs
    -0.60
     digitally
    -0.60
    POSITIVE LOGITS
    raz
    4.57
    rap
    1.35
    ripp
    1.30
    rac
    1.21
    riz
    1.17
    rad
    1.16
    ras
    1.15
    ram
    1.15
    raf
    1.15
    roc
    1.13
    Activations Density 0.015%

    No Known Activations