© 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 49152
    9. 8-RES_FS49152-JB
    11. 45846
    Prev
    Next
    INDEX
    Explanations

    mentions of vehicles or brands associated with automotive topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    659
    +0.18
    0.9%
    18
    +0.12
    0.6%
    743
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.18
    0.05
    18
    +0.12
    0.04
    743
    +0.10
    0.03
    Negative Logits
    ģĸ
    -0.74
    ãĥ¼ãĥĨ
    -0.67
    ãĤ´ãĥ³
    -0.67
    ¯¯
    -0.67
     Heist
    -0.65
    ãĥ³ãĤ¸
    -0.64
    dress
    -0.63
     tong
    -0.62
     deduction
    -0.61
    manship
    -0.61
    POSITIVE LOGITS
    ulture
    1.28
    oodoo
    1.25
    arsity
    1.23
    IRT
    1.18
    alky
    1.17
    ogue
    1.16
    isions
    1.16
    ampires
    1.12
    odka
    1.10
    andals
    1.05
    Activations Density 0.023%

    No Known Activations