© 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. 10359
    Prev
    Next
    INDEX
    Explanations

    words related to legal proceedings and charges

    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
    213
    +0.10
    0.5%
    394
    +0.10
    0.4%
    294
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    174
    +0.10
    0.03
    444
    +0.10
    0.02
    232
    +0.09
    0.03
    Negative Logits
     otherwise
    -0.64
     bed
    -0.62
     Kubrick
    -0.59
    SourceFile
    -0.59
     AVG
    -0.59
     regards
    -0.58
     tipped
    -0.57
    âĢİ
    -0.57
     hel
    -0.56
     Encyclopedia
    -0.56
    POSITIVE LOGITS
    achine
    1.29
    issions
    1.27
    peror
    1.22
    bourg
    1.14
    useum
    1.12
    poral
    1.11
    otional
    1.09
    obile
    1.08
    blance
    1.06
    ple
    1.05
    Activations Density 0.042%

    No Known Activations