© 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. 19777
    Prev
    Next
    INDEX
    Explanations

    words related to fear and disturbing situations

    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
    526
    +0.12
    0.6%
    649
    +0.11
    0.5%
    403
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.01
    403
    +0.11
    0.01
    526
    +0.11
    0.00
    Negative Logits
    ctions
    -0.81
    ction
    -0.78
     unification
    -0.76
    ufact
    -0.72
     sovere
    -0.71
    nas
    -0.70
    pled
    -0.68
    ãĥ¼ãĥ³
    -0.68
    pring
    -0.67
    enture
    -0.66
    POSITIVE LOGITS
     craw
    0.94
     Creep
    0.93
     creep
    0.92
    sylvania
    0.91
    crow
    0.88
    ritch
    0.86
     creeps
    0.84
    ulhu
    0.83
    yp
    0.83
    ening
    0.83
    Activations Density 0.023%

    No Known Activations