© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 45796
    Prev
    Next
    INDEX
    Explanations

    instances of the word "announce" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    418
    +0.13
    0.6%
    271
    +0.12
    0.5%
    721
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.13
    0.01
    411
    +0.12
    0.01
    101
    +0.10
    0.01
    Negative Logits
    acea
    -0.96
    Downloadha
    -0.81
    idences
    -0.78
    uther
    -0.77
    otherapy
    -0.76
    oscopic
    -0.76
    atural
    -0.71
    incial
    -0.70
    é¾įå
    -0.69
    isu
    -0.68
    POSITIVE LOGITS
    mare
    0.79
    ounce
    0.79
    rers
    0.76
    ment
    0.74
    ÙĴ
    0.72
    faced
    0.72
    Beast
    0.70
     Viet
    0.69
    rer
    0.68
     Seym
    0.68
    Activations Density 0.007%

    No Known Activations