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

    Neuronpedia

    APICircuit 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. 782
    Prev
    Next
    INDEX
    Explanations

    words related to politics and legal issues

    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
    488
    +0.11
    0.5%
    224
    +0.11
    0.5%
    470
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    350
    +0.11
    0.05
    488
    +0.11
    0.05
    121
    +0.10
    0.05
    Negative Logits
     psychologists
    -0.81
     Collider
    -0.74
     manuals
    -0.74
     crate
    -0.74
     theorists
    -0.73
    âĹ¼
    -0.72
    zsche
    -0.70
    Reviewer
    -0.70
     physicists
    -0.68
     catalog
    -0.68
    POSITIVE LOGITS
     Elections
    1.32
    election
    1.32
    bernatorial
    1.26
     constituency
    1.25
     mayoral
    1.21
     reelection
    1.19
     ballot
    1.17
     Voters
    1.17
     gubernatorial
    1.17
     runoff
    1.14
    Activations Density 0.413%

    No Known Activations