© 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. 18198
    Prev
    Next
    INDEX
    Explanations

    code snippets related to implementing cache mechanisms, memory management, and mathematical operations

    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
    447
    +0.16
    0.7%
    326
    +0.14
    0.6%
    334
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.03
    507
    +0.14
    0.03
    99
    +0.12
    0.03
    Negative Logits
     NRA
    -0.65
     coron
    -0.65
    idelines
    -0.63
     Loving
    -0.61
    hement
    -0.60
    ortmund
    -0.60
     Volvo
    -0.60
     Belle
    -0.60
     lieu
    -0.59
     Krug
    -0.59
    POSITIVE LOGITS
    000000
    1.41
    0000
    1.40
    0000000
    1.37
    0000000000000000
    1.32
    00000
    1.26
    00000000
    1.23
    9999
    1.15
    px
    1.06
    ÏĢ
    1.02
    999
    1.01
    Activations Density 0.105%

    No Known Activations