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. 487
    Prev
    Next
    INDEX
    Explanations

    factual inaccuracies or errors in information being discussed

    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
    216
    +0.13
    0.6%
    746
    +0.13
    0.6%
    481
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.04
    48
    +0.13
    0.03
    746
    +0.13
    0.02
    Negative Logits
    manship
    -0.72
     miscarriage
    -0.69
     DRAG
    -0.69
     Mutual
    -0.61
    20439
    -0.60
    ãģ¦
    -0.59
    ————
    -0.58
    dress
    -0.58
     Tone
    -0.57
     PowerPoint
    -0.57
    POSITIVE LOGITS
    icious
    1.09
    ados
    1.02
    ace
    0.99
    acies
    0.97
    ained
    0.96
    ading
    0.95
    icked
    0.95
    aded
    0.95
    atural
    0.94
    ches
    0.94
    Activations Density 0.067%

    No Known Activations