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. 306
    Prev
    Next
    INDEX
    Explanations

    business-related agreements and partnerships

    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
    101
    +0.10
    0.5%
    247
    +0.10
    0.5%
    528
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.04
    101
    +0.10
    0.05
    761
    +0.09
    0.03
    Negative Logits
    icators
    -0.71
    gypt
    -0.69
    ixel
    -0.67
    undown
    -0.67
    scan
    -0.67
    DoS
    -0.64
    fulness
    -0.63
     piled
    -0.63
    ickr
    -0.63
    pop
    -0.62
    POSITIVE LOGITS
     agreements
    1.51
     agreement
    1.38
     pact
    1.28
     Agreement
    1.24
    reements
    1.20
     partnership
    1.19
     partnerships
    1.14
     negotiations
    1.09
     deals
    1.07
     relationship
    1.06
    Activations Density 0.233%

    No Known Activations