© 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 24576
    9. 8-RES_FS24576-JB
    11. 6946
    Prev
    Next
    INDEX
    Explanations

    superlatives or extreme descriptions such as "most."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    190
    +0.12
    0.5%
    102
    +0.11
    0.5%
    553
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    190
    +0.12
    0.06
    553
    +0.11
    0.04
    194
    +0.10
    0.04
    Negative Logits
    halla
    -0.75
    instead
    -0.71
     Beans
    -0.69
     Sequence
    -0.66
     Translation
    -0.66
     Procedure
    -0.65
     Oaks
    -0.64
     Tags
    -0.64
    culosis
    -0.64
     Analytics
    -0.64
    POSITIVE LOGITS
     important
    1.20
     prominent
    1.15
     powerful
    1.15
     influential
    1.15
     lucrative
    1.11
     recognizable
    1.11
     prestigious
    1.11
     egregious
    1.08
     prolific
    1.05
     valuable
    1.02
    Activations Density 0.054%

    No Known Activations