© 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. 41509
    Prev
    Next
    INDEX
    Explanations

    adjectives and adverbs with specific endings

    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
    97
    +0.10
    0.4%
    49
    +0.10
    0.4%
    450
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    151
    +0.10
    0.03
    97
    +0.10
    0.02
    103
    +0.10
    0.02
    Negative Logits
     Riders
    -0.80
     Newsp
    -0.71
     Royals
    -0.70
     Chiefs
    -0.69
     RJ
    -0.69
    anche
    -0.69
     Seasons
    -0.68
     Pirate
    -0.67
     OC
    -0.66
    anca
    -0.66
    POSITIVE LOGITS
     inclined
    0.89
     Gleaming
    0.89
     minded
    0.88
    ãĤ©
    0.87
     speaking
    0.75
    ascus
    0.74
     opposed
    0.72
     diverse
    0.72
     indistinguishable
    0.71
    Ü
    0.70
    Activations Density 0.016%

    No Known Activations