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 3072
    5. 8-RES_FS3072-JB
    6. 830
    Prev
    Next
    INDEX
    Explanations

    phrases related to fans and fandom

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    63
    +0.12
    0.5%
    568
    +0.10
    0.4%
    376
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    568
    +0.12
    0.05
    63
    +0.10
    0.05
    32
    +0.09
    0.05
    Negative Logits
     bip
    -0.63
     Johnston
    -0.63
    innon
    -0.61
     diplom
    -0.61
     SOM
    -0.61
     Rog
    -0.60
     Alz
    -0.59
     Khe
    -0.58
     Neg
    -0.58
     Samar
    -0.58
    POSITIVE LOGITS
    atical
    1.23
    atics
    1.14
    boys
    1.12
    atically
    1.03
    fare
    1.02
    club
    1.00
    atic
    0.95
    haw
    0.95
    boy
    0.94
    ubs
    0.93
    Activations Density 0.554%

    No Known Activations