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. 1015
    Prev
    Next
    INDEX
    Explanations

    statements expressing certainty or strong affirmation

    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
    243
    +0.11
    0.5%
    151
    +0.10
    0.5%
    378
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    243
    +0.11
    0.06
    718
    +0.10
    0.06
    151
    +0.09
    0.06
    Negative Logits
    ingly
    -0.78
    awaru
    -0.74
    lay
    -0.72
    entary
    -0.72
    aily
    -0.70
    iously
    -0.68
    roups
    -0.65
    matically
    -0.65
    iless
    -0.65
    idas
    -0.65
    POSITIVE LOGITS
     qualifies
    0.78
     deserved
    0.70
     outnumbered
    0.69
     influenced
    0.69
     benefited
    0.67
     appreciated
    0.67
     NS
    0.66
     preferable
    0.66
     welcome
    0.64
     effected
    0.64
    Activations Density 0.551%

    No Known Activations