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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 16316
    Prev
    Next
    INDEX
    Explanations

    phrases related to neurochemicals and their effects on arousal and dopamine receptors

    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
    738
    +0.11
    0.5%
    481
    +0.11
    0.5%
    162
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    738
    +0.11
    0.04
    647
    +0.11
    0.03
    714
    +0.10
    0.03
    Negative Logits
    Marx
    -0.85
    selage
    -0.82
    Pie
    -0.80
    auga
    -0.78
     Ruins
    -0.76
     Spaces
    -0.76
     Buildings
    -0.76
     Geological
    -0.73
    izens
    -0.72
     Franchise
    -0.72
    POSITIVE LOGITS
     therapies
    1.49
     therapeutic
    1.46
     medications
    1.45
     therape
    1.44
     medication
    1.43
     chemotherapy
    1.38
     treatments
    1.37
     therapy
    1.36
     dosage
    1.35
     administered
    1.33
    Activations Density 0.480%

    No Known Activations