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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 9788
    Prev
    Next
    INDEX
    Explanations

    information related to television show episodes and seasons, particularly in the context of TV series storylines

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    447
    +0.12
    0.6%
    228
    +0.12
    0.5%
    651
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.12
    0.04
    228
    +0.12
    0.03
    76
    +0.11
    0.03
    Negative Logits
    ntil
    -0.92
    artisan
    -0.77
    hari
    -0.74
    minist
    -0.73
    erness
    -0.71
    jri
    -0.70
    emort
    -0.69
    punk
    -0.69
    merce
    -0.69
    umbai
    -0.68
    POSITIVE LOGITS
     finale
    1.01
     episode
    0.95
     aired
    0.93
     episodes
    0.92
     premiere
    0.92
     airing
    0.90
     premie
    0.88
     opener
    0.87
    Dialogue
    0.85
     Transcript
    0.85
    Activations Density 0.086%

    No Known Activations