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. 23066
    Prev
    Next
    INDEX
    Explanations

    the word "believe" and its variations in various 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
    74
    +0.10
    0.5%
    78
    +0.10
    0.5%
    640
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    74
    +0.10
    0.02
    78
    +0.10
    0.01
    272
    +0.10
    0.02
    Negative Logits
    soever
    -1.00
    DragonMagazine
    -0.78
    ener
    -0.75
    MAP
    -0.73
    YE
    -0.68
    IRO
    -0.67
    stroke
    -0.67
    OGR
    -0.66
    Ware
    -0.66
    arf
    -0.64
    POSITIVE LOGITS
     miracles
    1.12
    clus
    1.08
     preserving
    0.86
     independence
    0.84
     fairness
    0.83
     moderation
    0.82
     spite
    0.82
     openness
    0.81
     miraculous
    0.81
     equality
    0.79
    Activations Density 0.048%

    No Known Activations