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. 1667
    Prev
    Next
    INDEX
    Explanations

    information related to legal cases and investigations

    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
    87
    +0.27
    1.4%
    470
    +0.10
    0.5%
    121
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.27
    -0.07
    470
    +0.10
    0.07
    627
    +0.10
    0.05
    Negative Logits
     whilst
    -0.83
     civilisation
    -0.82
     fucking
    -0.74
     realise
    -0.73
     invention
    -0.70
    ´
    -0.69
     fuck
    -0.68
    Firstly
    -0.68
     infinitely
    -0.67
     splend
    -0.66
    POSITIVE LOGITS
     spokeswoman
    1.21
     spokesman
    1.11
     spokesperson
    0.85
     Grassley
    0.83
     attorney
    0.81
     NAACP
    0.81
     redacted
    0.81
     declined
    0.78
     attorneys
    0.78
     lawmaker
    0.78
    Activations Density 6.385%

    No Known Activations