© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 9381
    Prev
    Next
    INDEX
    Explanations

    phrases related to the Islamic State (ISIS)

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    288
    +0.16
    0.7%
    293
    +0.10
    0.5%
    733
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    733
    +0.16
    0.02
    293
    +0.10
    0.02
    752
    +0.10
    0.02
    Negative Logits
    drops
    -0.90
    éĹĺ
    -0.82
    bley
    -0.82
    erick
    -0.78
    butt
    -0.74
    ym
    -0.72
    buck
    -0.71
    nces
    -0.71
    vable
    -0.69
    GES
    -0.69
    POSITIVE LOGITS
     extremism
    1.04
     extremist
    1.01
     supremacists
    1.01
     supremacist
    0.99
     extremists
    0.99
     Jihad
    0.96
     Brotherhood
    0.93
     fundamentalist
    0.93
     Sharia
    0.91
    ophobic
    0.91
    Activations Density 0.016%

    No Known Activations