© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 1444
    Prev
    Next
    INDEX
    Explanations

    mentions of political figures and conflicts, as well as statements containing strong opinions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    +0.28
    1.4%
    756
    +0.12
    0.6%
    86
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.28
    -0.03
    487
    +0.12
    0.06
    86
    +0.10
    0.07
    Negative Logits
     Xuan
    -0.62
     peaks
    -0.59
     solo
    -0.59
    iatus
    -0.58
    Enlarge
    -0.58
     Variable
    -0.58
    toggle
    -0.58
    noticed
    -0.58
     Finder
    -0.57
    enhagen
    -0.57
    POSITIVE LOGITS
     disgrace
    1.07
     disrespect
    1.05
     unacceptable
    0.96
     irresponsible
    0.94
     hypocrisy
    0.93
     endanger
    0.89
     despicable
    0.89
     shameful
    0.88
     injustice
    0.87
     jeopard
    0.86
    Activations Density 1.579%

    No Known Activations