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. 11793
    Prev
    Next
    INDEX
    Explanations

    financial numbers and terms related to business transactions

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    481
    +0.17
    0.8%
    288
    +0.11
    0.5%
    599
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.06
    442
    +0.11
    0.04
    35
    +0.10
    0.04
    Negative Logits
    yip
    -0.68
    odox
    -0.63
    cially
    -0.62
    ĪĴ
    -0.61
    omorph
    -0.59
    ":"/
    -0.57
    itimate
    -0.57
    xit
    -0.57
    opic
    -0.56
     Silence
    -0.56
    POSITIVE LOGITS
     annually
    1.13
     apiece
    1.10
     per
    0.90
     compared
    0.87
     (£
    0.83
     equivalents
    0.82
     total
    0.79
     plus
    0.79
    /$
    0.79
     each
    0.77
    Activations Density 0.290%

    No Known Activations