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. 41246
    Prev
    Next
    INDEX
    Explanations

    instances of the word "intro" 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_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
    651
    +0.10
    0.5%
    710
    +0.10
    0.5%
    60
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    710
    +0.10
    0.01
    651
    +0.10
    0.01
    422
    +0.10
    0.01
    Negative Logits
     Millennium
    -0.76
     Ivory
    -0.65
     Sutherland
    -0.63
     Liang
    -0.62
     Shogun
    -0.60
     Pru
    -0.58
     Franks
    -0.58
     Laurent
    -0.58
     precincts
    -0.57
     Goods
    -0.57
    POSITIVE LOGITS
    spection
    1.80
    spective
    1.66
    verted
    1.59
    spect
    1.23
    verts
    1.23
    vert
    1.21
    version
    1.14
    verting
    1.06
    jection
    1.04
    hibited
    0.94
    Activations Density 0.004%

    No Known Activations