© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 25735
    Prev
    Next
    INDEX
    Explanations

    phrases that express a sense of appreciation or commentary on various situations

    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
    36
    +0.19
    0.9%
    393
    +0.11
    0.5%
    334
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.19
    0.02
    334
    +0.11
    0.01
    263
    +0.10
    0.01
    Negative Logits
     ¶
    -0.72
    76561
    -0.68
     ILCS
    -0.67
    é¾įå¥ij士
    -0.67
    apons
    -0.66
     Oswald
    -0.65
    sylvania
    -0.64
    \<
    -0.64
     withd
    -0.63
    Nov
    -0.62
    POSITIVE LOGITS
     coincidence
    1.07
     wonderful
    0.96
     bunch
    0.93
     lot
    0.93
     lovely
    0.93
     hypocr
    0.87
     marvelous
    0.86
     difference
    0.86
     heck
    0.85
     fantastic
    0.85
    Activations Density 0.009%

    No Known Activations