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. 23805
    Prev
    Next
    INDEX
    Explanations

    mentions of legal terms and concepts, particularly related to rights and permissions

    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
    408
    +0.11
    0.5%
    694
    +0.11
    0.5%
    392
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.11
    0.03
    395
    +0.11
    0.02
    298
    +0.10
    0.02
    Negative Logits
    estate
    -0.84
     âĢº
    -0.77
    ãĥĩ
    -0.75
     Rates
    -0.75
    adr
    -0.74
    geist
    -0.73
    ãĥ³ãĤ¸
    -0.73
    ahime
    -0.71
    ãĥ´
    -0.71
    estyle
    -0.69
    POSITIVE LOGITS
     interruption
    1.00
     permission
    0.97
     supervision
    0.94
     accompanying
    0.91
     pesky
    0.90
     intervention
    0.87
     exception
    0.86
     hesitation
    0.86
     doubt
    0.85
     consent
    0.85
    Activations Density 0.086%

    No Known Activations