© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 9311
    Prev
    Next
    INDEX
    Explanations

    references to geographical locations, specifically focusing on the south direction

    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
    496
    +0.17
    0.8%
    238
    +0.14
    0.6%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.17
    0.03
    596
    +0.14
    0.03
    496
    +0.11
    0.03
    Negative Logits
     MFT
    -0.70
     CLSID
    -0.69
    ãĤ¤ãĥĪ
    -0.69
     Soda
    -0.68
    elsen
    -0.66
    roma
    -0.65
    66666666
    -0.65
    ause
    -0.63
     "$:/
    -0.63
     Unch
    -0.63
    POSITIVE LOGITS
    wark
    1.19
    ward
    1.15
    wards
    0.95
    west
    0.93
    side
    0.92
    western
    0.91
    bound
    0.88
    wind
    0.86
    ampton
    0.85
    overs
    0.80
    Activations Density 0.019%

    No Known Activations