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. 8479
    Prev
    Next
    INDEX
    Explanations

    mentions of the Commerce Department and related activities

    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
    288
    +0.14
    0.7%
    71
    +0.12
    0.5%
    496
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.02
    508
    +0.12
    0.01
    467
    +0.11
    0.01
    Negative Logits
    etr
    -0.74
     MAT
    -0.67
    aze
    -0.66
    768
    -0.63
    arity
    -0.61
    otted
    -0.58
     TIT
    -0.58
    ractive
    -0.57
     Janeiro
    -0.57
    udeb
    -0.56
    POSITIVE LOGITS
    INESS
    0.97
    andise
    0.80
    halla
    0.78
    woman
    0.76
    merce
    0.75
    eka
    0.74
    cean
    0.73
    eers
    0.72
    ably
    0.70
     Commerce
    0.69
    Activations Density 0.024%

    No Known Activations