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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11379
    Prev
    Next
    INDEX
    Explanations

    statements related to public announcements or events

    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
    159
    +0.16
    0.7%
    186
    +0.11
    0.5%
    393
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.16
    0.10
    186
    +0.11
    0.06
    578
    +0.10
    0.04
    Negative Logits
    vae
    -0.81
    oller
    -0.75
    icons
    -0.73
    unk
    -0.68
    nets
    -0.67
    imal
    -0.66
    papers
    -0.66
    cery
    -0.64
    ickets
    -0.62
    worms
    -0.62
    POSITIVE LOGITS
     week
    1.16
     latest
    1.05
     year
    1.04
     month
    1.01
     morning
    0.87
     weekend
    0.85
     newfound
    0.79
     discrepancy
    0.79
     newest
    0.79
     particular
    0.77
    Activations Density 0.171%

    No Known Activations