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. 463
    Prev
    Next
    INDEX
    Explanations

    images or captions containing descriptions of mourning and events at a stadium

    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
    373
    +0.10
    0.4%
    299
    +0.09
    0.4%
    153
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.10
    0.01
    97
    +0.09
    0.01
    308
    +0.08
    0.01
    Negative Logits
     attrition
    -0.77
    bage
    -0.75
    nesia
    -0.70
    soType
    -0.67
    nexus
    -0.67
     sooner
    -0.65
    ¯
    -0.62
    ministic
    -0.61
     allele
    -0.61
    isition
    -0.61
    POSITIVE LOGITS
     Faces
    0.99
     Photos
    0.97
     Scenes
    0.93
     Gad
    0.80
    arnaev
    0.79
     ï
    0.76
     Countries
    0.73
     Top
    0.69
     Moments
    0.69
     Stru
    0.68
    Activations Density 0.035%

    No Known Activations