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. 21592
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Super" followed by a single digit number

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    167
    +0.12
    0.5%
    232
    +0.11
    0.5%
    128
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    493
    +0.11
    0.02
    727
    +0.09
    0.02
    Negative Logits
    gow
    -0.83
    assetsadobe
    -0.74
    ered
    -0.70
    OPLE
    -0.70
     Seym
    -0.69
    edIn
    -0.69
     guiActiveUn
    -0.67
    ...]
    -0.66
    asus
    -0.65
    76561
    -0.64
    POSITIVE LOGITS
    visors
    1.24
    visor
    1.19
    natural
    1.16
    nova
    1.14
    markets
    1.10
    charg
    1.06
    imposed
    1.05
    intend
    1.03
    bowl
    1.02
    intendent
    1.01
    Activations Density 0.013%

    No Known Activations