© 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 49152
    9. 8-RES_FS49152-JB
    11. 37367
    Prev
    Next
    INDEX
    Explanations

    references to the group Al Qaeda and its various iterations or affiliations

    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
    314
    +0.11
    0.5%
    213
    +0.11
    0.5%
    365
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    213
    +0.11
    0.02
    365
    +0.11
    0.02
    201
    +0.10
    0.02
    Negative Logits
     Lovecraft
    -0.87
    ãĥ¼ãĥĨãĤ£
    -0.87
    å£
    -0.80
     Vaugh
    -0.74
     Twain
    -0.69
     wip
    -0.68
     loft
    -0.67
    manship
    -0.67
     goodbye
    -0.67
     Seah
    -0.67
    POSITIVE LOGITS
     Jazeera
    1.38
     Qaeda
    1.11
    iyah
    1.08
    addin
    1.00
    aida
    0.98
    oha
    0.95
    Qaeda
    0.95
    manac
    0.94
    istar
    0.93
    gebra
    0.93
    Activations Density 0.013%

    No Known Activations