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. 292
    Prev
    Next
    INDEX
    Explanations

    descriptions of events or situations

    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
    373
    +0.13
    0.6%
    157
    +0.11
    0.5%
    619
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    619
    +0.13
    0.10
    157
    +0.11
    0.11
    484
    +0.10
    0.10
    Negative Logits
    aminer
    -0.68
    SHIP
    -0.65
    IDS
    -0.64
    ritic
    -0.62
    potion
    -0.59
    assembly
    -0.59
    rir
    -0.58
    yip
    -0.57
    gee
    -0.57
    plin
    -0.57
    POSITIVE LOGITS
    ocating
    1.41
    iances
    1.21
     kinds
    1.21
    igators
    1.19
     sorts
    1.18
    igator
    1.18
    usions
    1.15
    iance
    1.14
    ocated
    1.12
    uding
    1.11
    Activations Density 0.819%

    No Known Activations