© 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 3072
    9. 8-RES_FS3072-JB
    11. 739
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions or changes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    16
    +0.11
    0.5%
    206
    +0.10
    0.5%
    488
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    541
    +0.11
    0.09
    390
    +0.10
    0.07
    488
    +0.10
    0.08
    Negative Logits
    ierre
    -0.71
    cot
    -0.64
    itte
    -0.64
    RET
    -0.61
    POSE
    -0.60
    Forest
    -0.59
    IFT
    -0.58
    cipled
    -0.58
    enter
    -0.57
    atever
    -0.56
    POSITIVE LOGITS
     by
    1.25
     BY
    1.02
     herein
    0.97
     therein
    0.96
     aback
    0.88
     during
    0.82
     By
    0.78
    pez
    0.76
     pursuant
    0.75
     jointly
    0.74
    Activations Density 0.889%

    No Known Activations