Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 11324
    Prev
    Next
    INDEX
    Explanations

    dates and specific events or activities

    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
    640
    +0.32
    1.5%
    94
    +0.13
    0.6%
    756
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.32
    0.17
    94
    +0.13
    0.13
    615
    +0.10
    0.08
    Negative Logits
     dissu
    -0.86
     trusting
    -0.76
     endeav
    -0.74
    heit
    -0.73
     diseng
    -0.72
     ideal
    -0.67
     wanting
    -0.67
     fors
    -0.67
     unsuspecting
    -0.65
     hesitation
    -0.64
    POSITIVE LOGITS
    Contents
    0.84
    ³³³
    0.82
    ramid
    0.80
    PHOTOS
    0.78
    Prior
    0.75
    FY
    0.74
    ³³³³³³³³³³³³³³³³
    0.73
    However
    0.73
    Its
    0.73
    According
    0.72
    Activations Density 0.496%

    No Known Activations