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. 770
    Prev
    Next
    INDEX
    Explanations

    names of countries and their relationships or involvements

    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
    210
    +0.10
    0.4%
    658
    +0.10
    0.4%
    472
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    210
    +0.10
    0.03
    115
    +0.10
    0.04
    110
    +0.09
    0.03
    Negative Logits
    regor
    -0.69
    isks
    -0.69
    lance
    -0.68
    lesi
    -0.66
    urtles
    -0.65
    osures
    -0.64
    phies
    -0.64
    FILE
    -0.63
    Salt
    -0.63
    onyms
    -0.63
    POSITIVE LOGITS
     ratio
    1.34
     Ratio
    1.24
     combo
    1.23
     hybrid
    1.21
     ratios
    1.18
     hybrids
    1.16
     relations
    1.14
     relationship
    1.14
     partnership
    1.10
     dich
    1.09
    Activations Density 0.149%

    No Known Activations