© 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. 247
    Prev
    Next
    INDEX
    Explanations

    text related to political drama and confrontational interactions

    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
    68
    +0.10
    0.5%
    719
    +0.10
    0.5%
    575
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.10
    0.11
    630
    +0.10
    0.10
    719
    +0.10
    0.10
    Negative Logits
     seiz
    -0.79
    ãĥ¯ãĥ³
    -0.78
     disse
    -0.74
     snail
    -0.70
     Franch
    -0.70
     obser
    -0.68
    uckland
    -0.67
     dehuman
    -0.66
     scor
    -0.65
    icit
    -0.64
    POSITIVE LOGITS
    Ŀ
    1.56
    ¡
    1.24
    Ĵ
    1.05
    ¤
    1.02
    ī
    1.02
    ľ
    1.01
    Ķ
    0.99
    Ĩ
    0.99
    ¦
    0.98
    ĺ
    0.97
    Activations Density 0.240%

    No Known Activations