© 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 EvalsExportsinterp-engineNEWAPI 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. 24396
    Prev
    Next
    INDEX
    Explanations

    instances of the word "in" and other prepositions indicating location or context

    oai_token-act-pair · gpt-4o-miniTriggered 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
    447
    +0.16
    0.7%
    596
    +0.10
    0.5%
    322
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    22
    +0.16
    0.06
    123
    +0.10
    0.06
    322
    +0.10
    0.05
    Negative Logits
    0000000000000000
    -0.68
     appoint
    -0.65
    Iterator
    -0.62
     overwrite
    -0.62
    Password
    -0.61
     exited
    -0.60
    CVE
    -0.60
     +++
    -0.60
     %%
    -0.60
    Completed
    -0.60
    POSITIVE LOGITS
     academia
    1.22
    roads
    1.06
    effic
    0.99
     Europe
    0.98
     circles
    0.98
     vitro
    0.90
     society
    0.89
     Southeast
    0.89
    ordinate
    0.89
     relation
    0.88
    Activations Density 0.277%

    No Known Activations