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. 23511
    Prev
    Next
    INDEX
    Explanations

    keywords and phrases related to political events and entities

    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
    371
    +0.11
    0.5%
    550
    +0.10
    0.4%
    210
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    210
    +0.11
    0.04
    91
    +0.10
    0.06
    371
    +0.10
    0.05
    Negative Logits
    aples
    -0.67
    Required
    -0.66
    earchers
    -0.65
    ceivable
    -0.64
    uble
    -0.64
    eco
    -0.63
    Jewish
    -0.63
    anson
    -0.62
    romy
    -0.62
    isle
    -0.61
    POSITIVE LOGITS
     namely
    0.99
     Hugo
    0.81
     Charlotte
    0.79
    ãĤ©
    0.79
     Angus
    0.78
     Alvin
    0.77
     Polaris
    0.77
     Tiffany
    0.76
     RED
    0.76
     Siri
    0.75
    Activations Density 0.640%

    No Known Activations