© 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. 4-RES-JB
    11. 16411
    Prev
    Next
    INDEX
    Explanations

    mentions of the terrorist group Daesh

    oai_token-act-pair · gpt-3.5-turbo

    references to the terrorist group Daesh (ISIS)

    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
    hetti
    -0.95
    ously
    -0.79
    ULAR
    -0.75
    glers
    -0.73
    uez
    -0.66
    packing
    -0.65
    cius
    -0.65
    ulates
    -0.65
    ulating
    -0.64
    zes
    -0.64
    POSITIVE LOGITS
    rolet
    1.08
     Daesh
    0.96
    ignty
    0.79
    lehem
    0.79
    omething
    0.73
    illery
    0.72
    ensional
    0.72
    peak
    0.72
    izen
    0.68
    cha
    0.67
    Activations Density 0.027%

    No Known Activations