© 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 49152
    9. 8-RES_FS49152-JB
    11. 26562
    Prev
    Next
    INDEX
    Explanations

    actions related to reading and writing abilities, particularly challenges faced by individuals in these areas

    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
    468
    +0.14
    0.6%
    162
    +0.12
    0.5%
    215
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    468
    +0.14
    0.03
    70
    +0.12
    0.04
    505
    +0.11
    0.03
    Negative Logits
     exclus
    -0.73
    isSpecialOrderable
    -0.67
     Advis
    -0.67
    exclusive
    -0.66
     Outbreak
    -0.66
     assurances
    -0.65
    eks
    -0.65
     Insider
    -0.63
    isd
    -0.62
    iances
    -0.62
    POSITIVE LOGITS
     upright
    0.97
     properly
    0.94
     efficiently
    0.91
     anymore
    0.88
     comfortably
    0.88
     freely
    0.87
     outdoors
    0.86
     indoors
    0.86
     underwater
    0.84
     accurately
    0.80
    Activations Density 0.131%

    No Known Activations