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. 23218
    Prev
    Next
    INDEX
    Explanations

    future intentions or commitments expressed with "I'll."

    oai_token-act-pair · gpt-4o-miniTriggered 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
    354
    +0.10
    0.4%
    707
    +0.10
    0.4%
    583
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.10
    0.03
    354
    +0.10
    0.03
    665
    +0.09
    0.03
    Negative Logits
    Reviewer
    -0.68
    planet
    -0.66
    QUI
    -0.65
    cos
    -0.65
    Offline
    -0.65
    Therefore
    -0.62
    elo
    -0.62
    senal
    -0.62
    ARA
    -0.61
    Strong
    -0.60
    POSITIVE LOGITS
     gladly
    1.05
     gotta
    1.03
     probably
    0.92
     be
    0.90
     see
    0.89
     doubtless
    0.85
     never
    0.84
     get
    0.79
     happily
    0.79
     likely
    0.79
    Activations Density 0.032%

    No Known Activations