Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 2518
    Prev
    Next
    INDEX
    Explanations

    law enforcement-related terms and events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    538
    +0.10
    0.4%
    574
    +0.09
    0.4%
    509
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.10
    0.07
    628
    +0.09
    0.06
    146
    +0.09
    0.07
    Negative Logits
    xual
    -0.91
    opian
    -0.84
    yip
    -0.83
    igible
    -0.82
    ebted
    -0.78
    isted
    -0.74
    vironment
    -0.71
    eem
    -0.70
    piring
    -0.69
    bp
    -0.69
    POSITIVE LOGITS
     officers
    1.25
     departments
    1.21
     officer
    1.12
     Officers
    1.09
     commissioner
    1.05
     department
    1.03
     brutality
    1.03
     chiefs
    0.99
     academy
    0.98
     Commissioner
    0.97
    Activations Density 0.515%

    No Known Activations