© 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 12288
    9. 8-RES_FS12288-JB
    11. 9867
    Prev
    Next
    INDEX
    Explanations

    commands and instructions related to software development and file manipulation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    640
    +0.14
    0.6%
    393
    +0.13
    0.6%
    124
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.14
    0.05
    124
    +0.13
    0.05
    160
    +0.11
    0.04
    Negative Logits
     Emmy
    -0.78
     Norris
    -0.77
     Jasper
    -0.73
     Maher
    -0.71
     Phelps
    -0.68
     Burnett
    -0.68
     Hanson
    -0.67
     Irwin
    -0.67
     Yaz
    -0.67
     Eve
    -0.66
    POSITIVE LOGITS
    mode
    1.22
    tools
    1.22
    modules
    1.18
    generic
    1.17
    derived
    1.17
    setup
    1.17
    cache
    1.16
    based
    1.16
    san
    1.15
    sama
    1.15
    Activations Density 0.126%

    No Known Activations