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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 19531
    Prev
    Next
    INDEX
    Explanations

    strong verbs or actions related to politics and legislation

    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
    330
    +0.09
    0.4%
    251
    +0.09
    0.4%
    85
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    251
    +0.09
    0.06
    232
    +0.09
    0.06
    503
    +0.09
    0.04
    Negative Logits
     maker
    -0.65
     medium
    -0.64
     sacrament
    -0.63
     vul
    -0.63
     transitions
    -0.62
     examiner
    -0.62
     uniqueness
    -0.61
     disabilities
    -0.61
     identification
    -0.61
     nudity
    -0.60
    POSITIVE LOGITS
    udging
    1.04
    inged
    1.04
    ashing
    1.02
    isively
    1.01
    oused
    0.98
    ashed
    0.98
    aciously
    0.97
    umbled
    0.96
    agged
    0.94
    uously
    0.93
    Activations Density 0.174%

    No Known Activations