© 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. 45517
    Prev
    Next
    INDEX
    Explanations

    words related to actions or performance in a narrative context

    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
    538
    +0.10
    0.4%
    537
    +0.10
    0.4%
    43
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.10
    0.01
    702
    +0.10
    0.01
    43
    +0.10
    0.01
    Negative Logits
    GV
    -0.76
     Wikimedia
    -0.74
     DRAG
    -0.71
    ãĥ´
    -0.71
    Sham
    -0.70
    erenn
    -0.70
    getting
    -0.68
    arij
    -0.67
     consolidation
    -0.66
    fell
    -0.65
    POSITIVE LOGITS
     warn
    0.78
     approving
    0.78
     omin
    0.75
     voice
    0.74
     down
    0.72
     alerts
    0.72
     deaf
    0.71
     bells
    0.71
    ulate
    0.71
    ails
    0.68
    Activations Density 0.016%

    No Known Activations