© 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 24576
    9. 8-RES_FS24576-JB
    11. 19967
    Prev
    Next
    INDEX
    Explanations

    URL links contained in HTML anchor tags

    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
    381
    +0.14
    0.6%
    373
    +0.10
    0.5%
    496
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    381
    +0.14
    0.01
    701
    +0.10
    0.01
    160
    +0.09
    0.01
    Negative Logits
     stakes
    -0.67
     Pigs
    -0.65
     Lob
    -0.64
     Nightmares
    -0.62
     Quart
    -0.61
     Veg
    -0.61
     BP
    -0.60
    quished
    -0.60
     Dahl
    -0.59
     Chim
    -0.59
    POSITIVE LOGITS
    https
    1.30
    http
    1.26
    ../
    0.99
    //
    0.96
    javascript
    0.94
    {{
    0.93
    <?
    0.90
    www
    0.89
    Website
    0.88
    twitter
    0.87
    Activations Density 0.016%

    No Known Activations