© 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. 10341
    Prev
    Next
    INDEX
    Explanations

    phrases referring to legal issues or conflict resolution

    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
    362
    +0.19
    0.9%
    393
    +0.13
    0.6%
    481
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.19
    0.14
    393
    +0.13
    0.09
    608
    +0.12
    0.09
    Negative Logits
    quit
    -0.81
    acons
    -0.80
    nesty
    -0.77
    vantage
    -0.76
    interstitial
    -0.74
    byter
    -0.73
    dor
    -0.73
    isson
    -0.73
    jac
    -0.73
    aves
    -0.73
    POSITIVE LOGITS
     midst
    1.27
     equation
    1.25
     psyche
    1.14
     cosmos
    1.02
     workings
    1.01
     universe
    1.00
     relationship
    0.96
     world
    0.94
     dealings
    0.93
     environment
    0.93
    Activations Density 0.620%

    No Known Activations