© 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. 1266
    Prev
    Next
    INDEX
    Explanations

    short, brief statements or comments within a text

    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
    760
    +0.11
    0.5%
    302
    +0.10
    0.5%
    13
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    760
    +0.11
    0.03
    195
    +0.10
    0.02
    491
    +0.10
    0.03
    Negative Logits
    eworld
    -0.75
    ammed
    -0.73
    existent
    -0.71
    ardless
    -0.71
    aden
    -0.71
    Reilly
    -0.69
    uvian
    -0.69
    ankind
    -0.68
    "},"
    -0.68
    ð
    -0.68
    POSITIVE LOGITS
     note
    1.37
     caveat
    1.36
     disclaimer
    1.34
     caveats
    1.29
     reminder
    1.19
     recap
    1.13
     refres
    1.12
     clarification
    1.11
     Note
    1.10
     notes
    1.01
    Activations Density 0.136%

    No Known Activations