© 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. 10354
    Prev
    Next
    INDEX
    Explanations

    information on political debates and legislation surrounding health policy, specifically related to Obamacare, transgender rights, and abortion regulations

    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
    264
    +0.13
    0.6%
    575
    +0.11
    0.5%
    670
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    567
    +0.13
    0.05
    575
    +0.11
    0.08
    329
    +0.11
    0.05
    Negative Logits
     Zeit
    -0.71
    CONCLUS
    -0.68
    wow
    -0.67
    tons
    -0.67
     ;)
    -0.67
     partName
    -0.66
    Dialogue
    -0.65
    âĹ¼
    -0.65
    yssey
    -0.64
    rir
    -0.64
    POSITIVE LOGITS
     lawful
    1.03
     exempted
    1.00
     unlawful
    1.00
     lawfully
    0.99
     ineligible
    0.97
     federally
    0.94
     exemptions
    0.94
     prohibited
    0.92
     punishable
    0.90
     eligibility
    0.90
    Activations Density 1.106%

    No Known Activations