© 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. 11-RES-JB
    11. 1768
    Prev
    Next
    INDEX
    Explanations

    phrases related to legislative actions or political processes

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and references to legislative bodies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     slur
    -0.68
    BIL
    -0.68
    ENCY
    -0.67
     herbs
    -0.67
     nostalg
    -0.66
    ãĥŀ
    -0.64
    RAY
    -0.62
     Armen
    -0.62
    Anth
    -0.62
    emis
    -0.61
    POSITIVE LOGITS
     Senate
    1.49
    Senate
    1.31
     senate
    1.30
     governors
    1.26
     gubernatorial
    1.24
    bernatorial
    1.22
     Judiciary
    1.19
     governor
    1.16
     Assembly
    1.15
     House
    1.13
    Activations Density 0.104%

    No Known Activations