© 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. 2254
    Prev
    Next
    INDEX
    Explanations

    mentions of specific locations or events

    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
    314
    +0.14
    0.7%
    640
    +0.10
    0.5%
    115
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    115
    +0.14
    0.04
    152
    +0.10
    0.04
    147
    +0.10
    0.03
    Negative Logits
     replay
    -0.79
     thumbs
    -0.76
     coefficient
    -0.76
     headlights
    -0.76
     indicators
    -0.75
     regress
    -0.72
     "%
    -0.71
     revers
    -0.71
     elim
    -0.70
     recomp
    -0.70
    POSITIVE LOGITS
    based
    1.60
    area
    1.46
    centric
    1.28
    style
    1.21
    themed
    1.18
    sized
    1.14
    focused
    1.13
    bound
    1.08
    centered
    1.07
    quartered
    1.06
    Activations Density 0.044%

    No Known Activations