© 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 12288
    9. 8-RES_FS12288-JB
    11. 9771
    Prev
    Next
    INDEX
    Explanations

    phrases related to aviation incidents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    484
    +0.12
    0.6%
    741
    +0.12
    0.6%
    346
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    484
    +0.12
    0.04
    537
    +0.12
    0.04
    652
    +0.11
    0.04
    Negative Logits
    FINE
    -0.83
     Cumber
    -0.78
    âĶģ
    -0.78
    UGC
    -0.77
     Territories
    -0.69
    utenberg
    -0.69
    METHOD
    -0.69
    PUT
    -0.67
    esson
    -0.66
    ":""},{"
    -0.64
    POSITIVE LOGITS
     carrier
    1.10
     flown
    1.07
     airplanes
    1.07
     wreckage
    1.05
     hangar
    1.05
     prope
    1.02
     aircraft
    1.02
    planes
    1.01
     flying
    1.01
    plane
    1.01
    Activations Density 0.078%

    No Known Activations