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. 472
    Prev
    Next
    INDEX
    Explanations

    quotes or reported speech that express skepticism or doubt

    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
    544
    +0.10
    0.5%
    373
    +0.10
    0.4%
    368
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    598
    +0.10
    0.05
    203
    +0.10
    0.06
    208
    +0.09
    0.05
    Negative Logits
    rill
    -0.66
    ukong
    -0.64
    oker
    -0.61
    ãĤĮ
    -0.60
    world
    -0.59
    ng
    -0.58
    awan
    -0.58
    acca
    -0.57
    shore
    -0.57
    ammed
    -0.57
    POSITIVE LOGITS
     none
    1.15
     invariably
    0.95
     alas
    0.94
     mostly
    0.94
     unfortunately
    0.90
     mainly
    0.88
     overall
    0.88
    tons
    0.87
     ultimately
    0.86
     fortunately
    0.85
    Activations Density 0.274%

    No Known Activations