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 49152
    5. 8-RES_FS49152-JB
    6. 37088
    Prev
    Next
    INDEX
    Explanations

    terms related to actions, instructions, and their outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    481
    +0.57
    3.3%
    447
    +0.22
    1.3%
    480
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.57
    0.01
    266
    +0.22
    0.09
    481
    +0.17
    0.09
    Negative Logits
     millenn
    -0.67
     Moroc
    -0.60
    ogether
    -0.57
     Mehran
    -0.56
    luster
    -0.54
     adolesc
    -0.54
     Leban
    -0.54
     Smithsonian
    -0.54
     Vaugh
    -0.53
     stoked
    -0.53
    POSITIVE LOGITS
     doesnt
    0.77
     \'
    0.71
    /(
    0.63
    (_
    0.58
     [/
    0.58
     [+
    0.58
     fallacy
    0.58
     dont
    0.57
    [_
    0.56
     caus
    0.55
    Activations Density 0.849%

    No Known Activations