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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 431
    Prev
    Next
    INDEX
    Explanations

    mentions of political topics and legislation

    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
    31
    +0.13
    0.6%
    350
    +0.12
    0.5%
    573
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    350
    +0.13
    0.12
    31
    +0.12
    0.12
    224
    +0.11
    0.11
    Negative Logits
     civilisation
    -0.86
    tics
    -0.78
    zsche
    -0.75
     attribution
    -0.72
     genetics
    -0.72
    istically
    -0.71
     Shanghai
    -0.70
     analys
    -0.68
     bondage
    -0.67
     scouts
    -0.67
    POSITIVE LOGITS
    Senate
    1.62
    osponsors
    1.51
     Senate
    1.46
     senators
    1.40
    Senator
    1.40
    ospons
    1.40
     Caucus
    1.38
     senate
    1.38
    Republican
    1.36
    Republicans
    1.36
    Activations Density 2.686%

    No Known Activations