© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 725
    Prev
    Next
    INDEX
    Explanations

    references to Spider-Man and related characters and themes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    213
    +0.12
    0.6%
    738
    +0.12
    0.5%
    52
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.03
    738
    +0.12
    0.02
    366
    +0.11
    0.02
    Negative Logits
    brance
    -0.82
    ufact
    -0.82
    usable
    -0.71
    iband
    -0.69
    verson
    -0.68
    igated
    -0.65
    sts
    -0.65
     forc
    -0.64
    ãĥ¼ãĥ³
    -0.63
    estine
    -0.63
    POSITIVE LOGITS
    monkey
    1.09
    lings
    1.03
    web
    0.92
    Oak
    0.89
    Spider
    0.83
     webs
    0.83
    bite
    0.83
    tail
    0.82
     Bite
    0.82
    hook
    0.82
    Activations Density 0.127%

    No Known Activations