© 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 24576
    9. 8-RES_FS24576-JB
    11. 19957
    Prev
    Next
    INDEX
    Explanations

    legal and criminal activity-related phrases

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    699
    +0.14
    0.7%
    695
    +0.11
    0.5%
    587
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    699
    +0.14
    0.01
    711
    +0.11
    0.01
    404
    +0.10
    0.01
    Negative Logits
    EEK
    -0.69
    merce
    -0.69
    CLASS
    -0.67
    houses
    -0.67
    soDeliveryDate
    -0.66
    rir
    -0.66
    wich
    -0.66
    herer
    -0.66
    ĸļ
    -0.65
    herty
    -0.64
    POSITIVE LOGITS
    ysis
    1.32
    ogue
    1.07
    ity
    1.05
    arial
    0.96
    ities
    0.88
    ized
    0.83
    ogy
    0.79
     piety
    0.78
    eah
    0.76
    ibrary
    0.76
    Activations Density 0.023%

    No Known Activations