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. 2204
    Prev
    Next
    INDEX
    Explanations

    meteorological terms related to severe weather such as storms, winds, and coastal flooding

    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
    481
    +0.21
    1.1%
    87
    +0.16
    0.9%
    373
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    464
    +0.21
    0.11
    659
    +0.16
    0.11
    87
    +0.15
    -0.11
    Negative Logits
    )."
    -0.86
    ]."
    -0.79
    ."[
    -0.72
    .).
    -0.69
    .'"
    -0.66
    $.
    -0.66
    ).[
    -0.66
    .""
    -0.65
    '."
    -0.63
    .</
    -0.62
    POSITIVE LOGITS
    ':
    0.83
     Edit
    0.79
    !:
    0.74
    ?:
    0.70
    ¶
    0.67
     meanwhile
    0.67
     âĵĺ
    0.66
     Profile
    0.64
    ibliography
    0.60
    \":
    0.55
    Activations Density 6.990%

    No Known Activations