© 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. 10-RES-JB
    11. 3181
    Prev
    Next
    INDEX
    Explanations

    names, titles, and legal terms

    oai_token-act-pair · gpt-3.5-turbo

    names and titles related to people and roles in the justice system

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Duo
    -0.64
    ccess
    -0.61
     Flesh
    -0.61
    nings
    -0.60
     Fields
    -0.60
     Err
    -0.59
    xs
    -0.59
     Filter
    -0.58
    tc
    -0.57
     Pass
    -0.57
    POSITIVE LOGITS
     JUSTICE
    1.58
     MARK
    1.53
    WOOD
    1.51
     HOUSE
    1.49
     AMER
    1.49
     JOHN
    1.49
    BILL
    1.49
     SERVICE
    1.48
     ANGEL
    1.47
     FRI
    1.47
    Activations Density 0.277%

    No Known Activations