Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 887
    Prev
    Next
    INDEX
    Explanations

    technical terms related to computer science and politics

    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
    720
    +0.13
    0.6%
    481
    +0.12
    0.6%
    354
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.13
    0.06
    160
    +0.12
    0.07
    360
    +0.09
    0.05
    Negative Logits
     BS
    -0.67
     pulp
    -0.60
     Scenes
    -0.59
     Wit
    -0.57
    MSN
    -0.57
     Submit
    -0.53
     hect
    -0.52
     LF
    -0.52
     Bed
    -0.52
    zos
    -0.52
    POSITIVE LOGITS
    of
    1.34
     thereof
    1.10
    OF
    0.96
    oft
    0.89
     of
    0.85
    that
    0.85
    between
    0.84
    onto
    0.82
    against
    0.81
    Of
    0.81
    Activations Density 0.430%

    No Known Activations