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 24576
    5. 8-RES_FS24576-JB
    6. 905
    Prev
    Next
    INDEX
    Explanations

    French surnames

    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
    115
    +0.11
    0.5%
    704
    +0.11
    0.5%
    476
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.04
    373
    +0.11
    -0.02
    41
    +0.10
    0.03
    Negative Logits
     Skydragon
    -0.73
    ©¶æ
    -0.70
     Revenge
    -0.70
     Constantine
    -0.66
    ashtra
    -0.66
    phies
    -0.65
    \/\/
    -0.64
    ographers
    -0.64
    pher
    -0.62
     Egyptians
    -0.62
    POSITIVE LOGITS
    igham
    0.89
    jee
    0.72
    ailability
    0.70
    aii
    0.70
    gui
    0.65
    Dou
    0.65
    ahi
    0.65
    agara
    0.65
    vu
    0.65
    bas
    0.64
    Activations Density 0.142%

    No Known Activations