© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 42971
    Prev
    Next
    INDEX
    Explanations

    reporting phrases that indicate findings, claims, or observations from investigations and studies

    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
    373
    +0.12
    0.5%
    88
    +0.10
    0.5%
    756
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    536
    +0.12
    0.03
    274
    +0.10
    0.03
    275
    +0.10
    0.02
    Negative Logits
     totality
    -0.58
     Yam
    -0.56
     breakout
    -0.53
     TAMADRA
    -0.52
    iencies
    -0.52
     fixation
    -0.51
     dynasty
    -0.50
     dialog
    -0.50
     calibration
    -0.50
     Galactic
    -0.50
    POSITIVE LOGITS
    said
    0.75
    Reviewer
    0.73
     ().
    0.72
    told
    0.71
    ummer
    0.70
    !.
    0.69
    achus
    0.67
    ori
    0.66
    onel
    0.66
    essors
    0.65
    Activations Density 0.095%

    No Known Activations