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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 983
    Prev
    Next
    INDEX
    Explanations

    mentions of specific geographical locations and their corresponding statistics or reports

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    •"make sure" features in 768 and 1536

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    393
    +0.25
    1.3%
    362
    +0.24
    1.2%
    608
    +0.15
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.25
    0.37
    385
    +0.24
    0.27
    608
    +0.15
    0.28
    Negative Logits
     besides
    -0.86
     namely
    -0.83
     because
    -0.77
     without
    -0.77
    abel
    -0.74
     whilst
    -0.73
     owing
    -0.73
     with
    -0.71
     wherein
    -0.71
     whenever
    -0.71
    POSITIVE LOGITS
     aforementioned
    1.29
     same
    1.28
     latter
    1.28
     largest
    1.15
     United
    1.13
     highest
    1.11
     latest
    1.10
     likes
    1.10
     nation
    1.08
     smallest
    1.08
    Activations Density 3.275%

    No Known Activations