© 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. 9565
    Prev
    Next
    INDEX
    Explanations

    information related to political and economic topics

    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
    640
    +0.20
    1.0%
    160
    +0.16
    0.8%
    314
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.20
    0.07
    160
    +0.16
    0.07
    6
    +0.13
    0.07
    Negative Logits
     nightly
    -0.92
     downstream
    -0.82
     synerg
    -0.82
     casc
    -0.81
     commission
    -0.79
     netted
    -0.79
     agon
    -0.78
     tire
    -0.77
     weekend
    -0.77
     frontline
    -0.77
    POSITIVE LOGITS
    Answer
    1.88
    Well
    1.55
    Yes
    1.52
    Yeah
    1.50
    Absolutely
    1.46
    Question
    1.45
    ANS
    1.43
    Certainly
    1.38
    Actually
    1.37
    Reply
    1.35
    Activations Density 0.168%

    No Known Activations