© 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 3072
    9. 8-RES_FS3072-JB
    11. 33
    Prev
    Next
    INDEX
    Explanations

    phrases related to the use of items or skills for a specific purpose

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    60
    +0.11
    0.5%
    420
    +0.10
    0.5%
    285
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.08
    60
    +0.10
    0.06
    593
    +0.10
    0.05
    Negative Logits
    bernatorial
    -0.74
    iatus
    -0.69
    ufact
    -0.66
    boxing
    -0.65
    odiac
    -0.65
    inguished
    -0.65
    ukemia
    -0.64
    achelor
    -0.63
    emetery
    -0.63
    etsk
    -0.63
    POSITIVE LOGITS
     techniques
    0.96
     pseudonym
    0.94
     sparing
    0.92
     borrowed
    0.91
     levers
    0.90
     technique
    0.88
     leverage
    0.87
     gimm
    0.86
     imagination
    0.86
     strengths
    0.85
    Activations Density 1.501%

    No Known Activations