© 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. 559
    Prev
    Next
    INDEX
    Explanations

    terms related to software development processes and tools

    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
    709
    +0.15
    0.7%
    512
    +0.10
    0.4%
    477
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    709
    +0.15
    0.05
    237
    +0.10
    0.04
    349
    +0.09
    0.03
    Negative Logits
     extract
    -0.62
     fractions
    -0.57
     Levi
    -0.56
     Read
    -0.54
     cann
    -0.54
     gorilla
    -0.54
     Cot
    -0.54
     Barg
    -0.53
     implicit
    -0.53
     stake
    -0.52
    POSITIVE LOGITS
    ,-
    0.93
    ,
    0.86
    KEN
    0.85
    Merit
    0.81
    )=(
    0.80
    ,,
    0.79
    maxwell
    0.78
    ,...
    0.77
    \<
    0.75
    ãĥ»
    0.75
    Activations Density 0.309%

    No Known Activations