© 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. 9869
    Prev
    Next
    INDEX
    Explanations

    steps or processes explained in a structured order

    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
    184
    +0.09
    0.4%
    481
    +0.09
    0.4%
    182
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    119
    +0.09
    0.04
    641
    +0.09
    0.05
    255
    +0.08
    0.04
    Negative Logits
    elaide
    -0.82
    idays
    -0.77
    oubted
    -0.72
    olls
    -0.71
    mosp
    -0.71
    aucas
    -0.70
     Voices
    -0.69
    nesty
    -0.69
     Mour
    -0.69
    crim
    -0.68
    POSITIVE LOGITS
     tricks
    1.17
     tutorial
    1.17
     instructions
    1.16
     routines
    1.15
     tutorials
    1.15
     technique
    1.15
    Instruct
    1.11
     manual
    1.10
     techniques
    1.08
     procedure
    1.07
    Activations Density 0.933%

    No Known Activations