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 49152
    5. 8-RES_FS49152-JB
    6. 26150
    Prev
    Next
    INDEX
    Explanations

    conditional statements or phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    86
    +0.10
    0.5%
    529
    +0.09
    0.4%
    300
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    529
    +0.10
    0.04
    300
    +0.09
    0.04
    715
    +0.08
    0.04
    Negative Logits
     Deity
    -0.72
    chant
    -0.71
    Eye
    -0.70
    ryu
    -0.68
    oves
    -0.66
    ãĥŃ
    -0.65
    ights
    -0.64
    pmwiki
    -0.63
    iery
    -0.62
    hig
    -0.61
    POSITIVE LOGITS
     successful
    0.98
     unsuccessful
    0.85
     elected
    0.84
     anything
    0.75
     policymakers
    0.75
     unchecked
    0.72
     convicted
    0.71
     you
    0.71
    fy
    0.70
     anyone
    0.69
    Activations Density 0.078%

    No Known Activations