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. 23592
    Prev
    Next
    INDEX
    Explanations

    references to transportation infrastructure and road conditions

    oai_token-act-pair · gpt-4o-miniTriggered 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
    325
    +0.11
    0.5%
    515
    +0.11
    0.5%
    41
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    325
    +0.11
    0.07
    41
    +0.11
    0.05
    537
    +0.10
    0.05
    Negative Logits
    azz
    -0.72
     Werewolf
    -0.71
    untu
    -0.71
    nikov
    -0.70
    ulz
    -0.68
    ONY
    -0.67
     cand
    -0.66
     mood
    -0.65
    illian
    -0.64
     diapers
    -0.63
    POSITIVE LOGITS
     highways
    1.28
     corridor
    1.28
    Route
    1.25
     tunnels
    1.21
     routes
    1.21
     lanes
    1.19
     crossings
    1.19
     corridors
    1.16
     Route
    1.15
     roads
    1.15
    Activations Density 0.430%

    No Known Activations