© 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 768
    9. 8-RES_FS768-JB
    11. 741
    Prev
    Next
    INDEX
    Explanations

    comparative phrases indicating greater importance or intensity

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    375
    +0.11
    0.5%
    539
    +0.10
    0.5%
    538
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    655
    +0.11
    0.10
    539
    +0.10
    0.08
    692
    +0.10
    0.06
    Negative Logits
    ALE
    -0.72
     largeDownload
    -0.72
    EMENT
    -0.69
    olitan
    -0.69
    ISSION
    -0.69
    cu
    -0.67
    uto
    -0.66
    imity
    -0.65
    aird
    -0.64
    avi
    -0.63
    POSITIVE LOGITS
     usual
    1.18
     anything
    1.03
     anybody
    0.85
     any
    0.84
     anticipated
    0.83
    average
    0.81
     anyone
    0.79
     mere
    0.79
     anywhere
    0.79
     expected
    0.78
    Activations Density 2.397%

    No Known Activations