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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 3-RES-JB
    11. 15091
    Prev
    Next
    INDEX
    Explanations

    references to the terrorist organization Daesh/ISIS and related activities

    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 @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hetti
    -0.81
    ships
    -0.77
    Wheel
    -0.74
    packing
    -0.73
    felt
    -0.70
    maid
    -0.69
    drm
    -0.69
    Reviewer
    -0.69
    hend
    -0.67
    odcast
    -0.66
    POSITIVE LOGITS
     Daesh
    1.25
     terrorists
    0.89
    ignty
    0.87
     militants
    0.84
    ISIS
    0.78
     terrorism
    0.77
    anie
    0.76
    merga
    0.76
     insurgents
    0.76
     ISIL
    0.76
    Activations Density 0.006%

    No Known Activations