© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 7079
    Prev
    Next
    INDEX
    Explanations

    instances of specific names like "Christie" and "Trump" along with terms related to politics and law

    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
    756
    +0.12
    0.5%
    128
    +0.11
    0.5%
    392
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    128
    +0.12
    0.05
    393
    +0.11
    0.05
    284
    +0.11
    0.04
    Negative Logits
    士
    -0.82
    ulhu
    -0.75
    GMT
    -0.72
     Explosion
    -0.71
    _.
    -0.71
    Planet
    -0.69
    é»Ĵ
    -0.69
    sbm
    -0.68
    tten
    -0.68
    punk
    -0.68
    POSITIVE LOGITS
     encouragement
    1.21
     encour
    1.05
     assist
    1.03
     supportive
    1.00
     counsel
    0.97
     assistance
    0.97
     assisting
    0.94
     aid
    0.93
     nurturing
    0.91
     endorsement
    0.90
    Activations Density 0.264%

    No Known Activations