© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 19848
    Prev
    Next
    INDEX
    Explanations

    mentions of locations and places

    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
    87
    +0.31
    1.7%
    756
    +0.14
    0.8%
    481
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    384
    +0.31
    0.05
    393
    +0.14
    0.07
    694
    +0.12
    0.04
    Negative Logits
    ptoms
    -0.67
    iquette
    -0.64
    ocent
    -0.63
    Definition
    -0.61
    BIP
    -0.61
    uid
    -0.59
    ãĤ£
    -0.56
    ophobia
    -0.55
    IGHTS
    -0.55
    DATA
    -0.55
    POSITIVE LOGITS
     etc
    0.73
     Scandinav
    0.71
     Azerb
    0.70
     Tel
    0.68
     Jamaica
    0.67
     Taiwan
    0.66
     Sahara
    0.65
     assorted
    0.64
     universities
    0.64
     Guam
    0.64
    Activations Density 0.414%

    No Known Activations