© 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. 23039
    Prev
    Next
    INDEX
    Explanations

    terms related to justifications or rationalizations for actions and behaviors

    oai_token-act-pair · gpt-4o-miniTriggered 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
    516
    +0.11
    0.5%
    681
    +0.10
    0.5%
    379
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    705
    +0.11
    0.01
    324
    +0.10
    0.01
    114
    +0.10
    0.01
    Negative Logits
    yss
    -0.93
    itals
    -0.82
    acid
    -0.81
    atform
    -0.79
    arya
    -0.75
    raid
    -0.74
    ecycle
    -0.74
    duino
    -0.73
    erial
    -0.73
    imble
    -0.73
    POSITIVE LOGITS
     excuse
    1.22
     excuses
    1.17
     explanations
    0.88
     justifying
    0.86
    abl
    0.81
     explanation
    0.80
     justification
    0.79
     why
    0.75
     overlooking
    0.75
     WHY
    0.74
    Activations Density 0.019%

    No Known Activations