© 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. 45676
    Prev
    Next
    INDEX
    Explanations

    the repeated use of the word "have" in different contexts

    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
    566
    +0.13
    0.6%
    7
    +0.10
    0.5%
    287
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.13
    0.04
    7
    +0.10
    0.02
    161
    +0.10
    0.02
    Negative Logits
    weed
    -0.76
    Discuss
    -0.74
    catentry
    -0.69
    TG
    -0.66
    ove
    -0.62
     Returning
    -0.62
    airs
    -0.62
    AH
    -0.62
    IM
    -0.61
    acs
    -0.60
    POSITIVE LOGITS
     been
    1.08
    been
    0.95
     gotta
    0.95
     seen
    0.94
     heard
    0.79
     gotten
    0.78
     chosen
    0.76
    Tube
    0.74
    seen
    0.73
     eaten
    0.72
    Activations Density 0.041%

    No Known Activations