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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 45955
    Prev
    Next
    INDEX
    Explanations

    instances of the letter "l" in the text

    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
    659
    +0.12
    0.6%
    18
    +0.10
    0.5%
    556
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.12
    0.03
    137
    +0.10
    0.02
    556
    +0.10
    0.02
    Negative Logits
    éĹĺ
    -0.70
    ãĥ¯ãĥ³
    -0.69
    chell
    -0.66
    ãĥ¼ãĥĨãĤ£
    -0.64
    lished
    -0.64
    OLOGY
    -0.62
    hower
    -0.62
    ãĥ¼ãĤ¯
    -0.59
     Credit
    -0.59
     Primal
    -0.59
    POSITIVE LOGITS
    adders
    1.39
    ibrarian
    1.33
    ibr
    1.27
    ongh
    1.17
    onel
    1.16
    idd
    1.15
    izards
    1.14
    ivery
    1.13
    otto
    1.13
    ugs
    1.12
    Activations Density 0.011%

    No Known Activations