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

    Neuronpedia

    APICircuit 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. 486
    Prev
    Next
    INDEX
    Explanations

    sports-related terms, specifically related to golf, and medical or law-related terms

    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
    481
    +0.23
    1.1%
    447
    +0.12
    0.5%
    459
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.23
    0.06
    312
    +0.12
    0.04
    575
    +0.12
    0.06
    Negative Logits
    oulos
    -0.77
    usalem
    -0.76
    inas
    -0.72
    daq
    -0.71
    ħĭ
    -0.71
    isitions
    -0.68
     Editors
    -0.68
    borough
    -0.65
     sidx
    -0.65
    racuse
    -0.63
    POSITIVE LOGITS
     requiring
    0.89
     invented
    0.87
     whose
    0.81
     whereby
    0.80
     commonly
    0.78
     devised
    0.78
     prized
    0.77
     meant
    0.75
     designed
    0.75
    horse
    0.72
    Activations Density 0.266%

    No Known Activations