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 24576
    5. 8-RES_FS24576-JB
    6. 23195
    Prev
    Next
    INDEX
    Explanations

    terms related to physical conditions and treatment contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.38
    2.1%
    678
    +0.09
    0.5%
    374
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.38
    -0.02
    570
    +0.09
    0.02
    678
    +0.08
    0.06
    Negative Logits
     sqor
    -0.70
    pmwiki
    -0.66
    etheless
    -0.64
    ETF
    -0.64
    CONCLUS
    -0.59
    Conclusion
    -0.56
    ologies
    -0.55
     Conclusion
    -0.55
     academia
    -0.54
    è¦ļéĨĴ
    -0.53
    POSITIVE LOGITS
     bathing
    0.66
     menstru
    0.65
     circumcised
    0.63
     chewing
    0.60
     feminine
    0.60
     toilet
    0.60
     breasts
    0.59
     erection
    0.57
     breast
    0.57
     greeting
    0.57
    Activations Density 1.278%

    No Known Activations