© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 37407
    Prev
    Next
    INDEX
    Explanations

    terms related to legal suspicions and charges

    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
    197
    +0.11
    0.5%
    321
    +0.10
    0.5%
    260
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    268
    +0.11
    0.02
    386
    +0.10
    0.02
    260
    +0.10
    0.02
    Negative Logits
    ongh
    -0.73
    vantage
    -0.72
     obser
    -0.71
    reen
    -0.67
    ********************************
    -0.67
    option
    -0.67
    aud
    -0.66
    options
    -0.65
    riel
    -0.65
    soDeliveryDate
    -0.64
    POSITIVE LOGITS
     misdem
    0.91
     felony
    0.87
     ranging
    0.81
     violations
    0.80
     unrelated
    0.79
     stemming
    0.78
     conspiring
    0.78
     treason
    0.77
     misdemeanor
    0.77
     charges
    0.76
    Activations Density 0.031%

    No Known Activations