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. 41376
    Prev
    Next
    INDEX
    Explanations

    references to tools, components, and resources related to equipment and technology

    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
    545
    +0.12
    0.5%
    289
    +0.10
    0.5%
    280
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    545
    +0.12
    0.02
    227
    +0.10
    0.02
    504
    +0.10
    0.02
    Negative Logits
    asio
    -0.74
     Vision
    -0.66
    Args
    -0.66
    rique
    -0.64
    osi
    -0.62
    liest
    -0.61
    SIGN
    -0.61
    ALLY
    -0.60
     exception
    -0.60
    Harris
    -0.59
    POSITIVE LOGITS
    paces
    1.38
    hip
    1.31
    pace
    1.28
    hips
    1.26
    ystem
    1.22
    mith
    1.21
    poons
    1.17
    ource
    1.17
    pread
    1.12
    heet
    1.10
    Activations Density 0.057%

    No Known Activations