© 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. 18859
    Prev
    Next
    INDEX
    Explanations

    mentions of advocacy groups or organizations

    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
    679
    +0.18
    0.8%
    27
    +0.11
    0.5%
    114
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.18
    0.06
    114
    +0.11
    0.04
    714
    +0.10
    0.04
    Negative Logits
    Phones
    -0.66
    stones
    -0.65
    dri
    -0.64
    elin
    -0.62
     Kardashian
    -0.61
     Chronicles
    -0.59
    Shock
    -0.59
    boards
    -0.58
    ultimate
    -0.58
    Delivery
    -0.57
    POSITIVE LOGITS
    hips
    1.07
     affiliated
    1.07
    ystem
    1.04
     vying
    0.98
    chool
    0.96
    folk
    0.95
    ourcing
    0.94
     sponsoring
    0.92
    paces
    0.91
    etter
    0.90
    Activations Density 0.154%

    No Known Activations