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 1536
    5. 8-RES_FS1536-JB
    6. 928
    Prev
    Next
    INDEX
    Explanations

    statistics related to sports

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    316
    +0.13
    0.6%
    627
    +0.10
    0.4%
    707
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    316
    +0.13
    0.07
    627
    +0.10
    0.05
    512
    +0.09
    0.07
    Negative Logits
     Ange
    -0.57
    dayName
    -0.55
    PUT
    -0.55
    externalActionCode
    -0.55
    SpaceEngineers
    -0.54
    verse
    -0.52
     live
    -0.52
    uyomi
    -0.51
     Provided
    -0.50
     Bend
    -0.50
    POSITIVE LOGITS
     39
    1.07
     38
    1.02
     33
    1.01
     31
    1.01
     36
    1.01
     28
    1.00
     27
    1.00
     37
    0.99
     29
    0.99
     17
    0.99
    Activations Density 1.411%

    No Known Activations