© 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. 770
    Prev
    Next
    INDEX
    Explanations

    references to specific locations about a certain region or situation

    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
    652
    +0.11
    0.5%
    757
    +0.10
    0.4%
    186
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    652
    +0.11
    0.04
    757
    +0.10
    0.03
    198
    +0.09
    0.03
    Negative Logits
     waiver
    -0.63
     Nanto
    -0.62
     Ventura
    -0.62
     Higgins
    -0.62
    eer
    -0.59
    eers
    -0.58
     Bast
    -0.56
     Russo
    -0.55
     Slayer
    -0.55
     McMahon
    -0.54
    POSITIVE LOGITS
    abouts
    1.91
    upon
    1.26
    tics
    0.99
    with
    0.96
    tical
    0.93
    comings
    0.91
     guiActiveUn
    0.85
    from
    0.82
    after
    0.80
     illegally
    0.79
    Activations Density 0.100%

    No Known Activations