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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 41633
    Prev
    Next
    INDEX
    Explanations

    parentheses and their contents

    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
    314
    +0.21
    1.0%
    640
    +0.19
    0.9%
    160
    +0.15
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.21
    0.07
    314
    +0.19
    0.02
    73
    +0.15
    0.05
    Negative Logits
     lag
    -0.75
     charism
    -0.71
     glare
    -0.70
     tam
    -0.70
     react
    -0.70
     instinct
    -0.69
     creep
    -0.69
     intelligent
    -0.69
     lull
    -0.68
     wre
    -0.66
    POSITIVE LOGITS
    albeit
    1.51
    although
    1.41
    including
    1.40
    which
    1.35
    excluding
    1.35
    pictured
    1.33
    along
    1.31
    though
    1.29
    mostly
    1.26
    see
    1.25
    Activations Density 0.104%

    No Known Activations