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. 23774
    Prev
    Next
    INDEX
    Explanations

    terms related to frameworks and connections in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    686
    +0.11
    0.5%
    324
    +0.10
    0.4%
    506
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.11
    0.05
    288
    +0.10
    0.05
    324
    +0.10
    0.04
    Negative Logits
    ancies
    -0.71
    ynski
    -0.71
    qus
    -0.69
    tions
    -0.65
     motions
    -0.62
    Downloadha
    -0.61
     Cancel
    -0.61
    orks
    -0.60
     Kingdoms
    -0.60
    @#&
    -0.59
    POSITIVE LOGITS
     reminder
    0.89
    piece
    0.89
     gateway
    0.86
     for
    0.84
    wark
    0.82
     fodder
    0.79
    imum
    0.78
    bridge
    0.76
    ifier
    0.76
     unto
    0.75
    Activations Density 0.229%

    No Known Activations