Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 43135
    Prev
    Next
    INDEX
    Explanations

    instances of the word "ut" in various forms

    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
    373
    +0.14
    0.7%
    284
    +0.12
    0.6%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.14
    0.01
    373
    +0.12
    -0.00
    406
    +0.10
    0.01
    Negative Logits
     compr
    -0.75
     trave
    -0.68
    Completed
    -0.67
     accomp
    -0.65
     coasts
    -0.65
    zona
    -0.62
    axter
    -0.61
     challeng
    -0.61
     contrace
    -0.59
    SPONSORED
    -0.58
    POSITIVE LOGITS
    cheon
    1.18
    tle
    1.03
    aneous
    0.93
    nik
    0.90
    ierrez
    0.87
    inous
    0.79
    chen
    0.79
    rition
    0.78
    ivated
    0.78
    ting
    0.78
    Activations Density 0.005%

    No Known Activations