© 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. 9335
    Prev
    Next
    INDEX
    Explanations

    financial and economic terms or concepts

    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
    331
    +0.10
    0.4%
    734
    +0.10
    0.4%
    158
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.10
    0.03
    102
    +0.10
    0.02
    734
    +0.09
    0.02
    Negative Logits
    arta
    -0.90
    abases
    -0.83
    aucas
    -0.77
    elle
    -0.75
    anwhile
    -0.74
    £ı
    -0.74
    Introduced
    -0.73
    afia
    -0.72
    olon
    -0.70
    udi
    -0.70
    POSITIVE LOGITS
     fruitful
    0.94
     disastrous
    0.88
     invaluable
    0.88
     quite
    0.85
     counterproductive
    0.81
     costly
    0.80
     useful
    0.78
     moot
    0.76
     hilarious
    0.76
     decisive
    0.75
    Activations Density 0.076%

    No Known Activations