© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10833
    Prev
    Next
    INDEX
    Explanations

    names of people or organizations involved in politics, business, or 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
    556
    +0.12
    0.5%
    459
    +0.11
    0.5%
    139
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    556
    +0.12
    0.07
    401
    +0.11
    0.07
    475
    +0.11
    0.05
    Negative Logits
    Ö¼
    -0.52
    ãĤ°
    -0.52
     anyway
    -0.51
     LOC
    -0.51
     mah
    -0.50
     anyways
    -0.49
     Kafka
    -0.49
     AZ
    -0.49
     liking
    -0.48
     Bellev
    -0.48
    POSITIVE LOGITS
     respectively
    1.31
     were
    1.20
     are
    1.18
    were
    1.06
     weren
    1.00
     converge
    0.97
     have
    0.91
     attest
    0.91
     aren
    0.91
     seem
    0.90
    Activations Density 0.455%

    No Known Activations