© 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. 19952
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to entertainment, sports, and places

    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
    326
    +0.10
    0.5%
    662
    +0.10
    0.5%
    376
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.10
    0.01
    343
    +0.10
    0.01
    628
    +0.10
    0.01
    Negative Logits
    eers
    -0.78
    matic
    -0.75
     Trafford
    -0.70
    ership
    -0.69
    aminer
    -0.69
    regor
    -0.68
    papers
    -0.68
    ARY
    -0.67
    omez
    -0.65
    geist
    -0.64
    POSITIVE LOGITS
    ille
    1.12
    ided
    1.10
    veland
    1.04
    zz
    1.03
    hma
    0.97
    igslist
    0.94
    very
    0.94
    kes
    0.93
    zzi
    0.91
    iding
    0.90
    Activations Density 0.019%

    No Known Activations