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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 17510
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical descriptions or personal possessions

    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
    712
    +0.10
    0.5%
    608
    +0.10
    0.4%
    194
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.10
    0.05
    362
    +0.10
    0.05
    712
    +0.09
    0.03
    Negative Logits
    tsy
    -0.81
    tch
    -0.73
    Gun
    -0.71
    roup
    -0.69
     Slate
    -0.65
    basketball
    -0.65
     Patton
    -0.65
    amiya
    -0.64
     Volunte
    -0.64
    rette
    -0.64
    POSITIVE LOGITS
     usefulness
    1.15
     predecessor
    1.12
     origins
    1.12
    ELF
    1.12
     contents
    1.10
     predecessors
    1.10
     namesake
    1.08
     merits
    1.07
     creator
    1.04
     creators
    1.04
    Activations Density 0.156%

    No Known Activations