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 49152
    5. 8-RES_FS49152-JB
    6. 26042
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the subject of discussion or topics being addressed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    447
    +0.13
    0.6%
    393
    +0.09
    0.4%
    723
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    687
    +0.13
    0.01
    348
    +0.09
    0.01
    723
    +0.09
    0.01
    Negative Logits
    DragonMagazine
    -0.94
    SPONSORED
    -0.88
     CLSID
    -0.78
    OGR
    -0.77
    nea
    -0.76
     entirety
    -0.75
    psons
    -0.74
    SON
    -0.73
    SPA
    -0.73
    SCP
    -0.72
    POSITIVE LOGITS
     preserving
    0.91
     protecting
    0.85
     educating
    0.85
     aesthetics
    0.84
     maximizing
    0.81
     ensuring
    0.77
     keeping
    0.76
     improving
    0.76
     creating
    0.76
     avoiding
    0.76
    Activations Density 0.019%

    No Known Activations