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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 846
    Prev
    Next
    INDEX
    Explanations

    phrases related to political, economic, or social issues and challenges, particularly focusing on problems and crises

    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
    756
    +0.17
    0.8%
    1
    +0.11
    0.5%
    410
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.17
    -0.02
    102
    +0.11
    0.13
    150
    +0.11
    0.07
    Negative Logits
     slots
    -0.89
     nods
    -0.81
    urai
    -0.71
     salsa
    -0.71
    çīĪ
    -0.69
     choir
    -0.69
     desks
    -0.69
     chapel
    -0.68
     specs
    -0.67
    racuse
    -0.67
    POSITIVE LOGITS
     exacerbated
    1.42
     caused
    1.37
     perpetrated
    1.30
     plag
    1.29
     inflicted
    1.28
     worsened
    1.26
     plague
    1.19
     endemic
    1.18
     severity
    1.16
     occurring
    1.13
    Activations Density 3.009%

    No Known Activations