© 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. 1263
    Prev
    Next
    INDEX
    Explanations

    beneficial outcomes or situations for different entities or individuals

    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
    311
    +0.12
    0.5%
    289
    +0.11
    0.5%
    352
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    555
    +0.12
    0.03
    380
    +0.11
    0.02
    48
    +0.10
    0.02
    Negative Logits
     Dragonbound
    -0.83
     traumatic
    -0.69
    eer
    -0.67
     unconventional
    -0.65
     fracturing
    -0.62
    LY
    -0.62
     Exhibit
    -0.62
     endlessly
    -0.61
     Elk
    -0.60
    tone
    -0.60
    POSITIVE LOGITS
    avement
    1.19
    isance
    1.16
    irmed
    1.13
    avored
    1.09
    ctions
    1.07
    ighed
    1.03
    ÃŁ
    1.03
    aviour
    0.99
    icial
    0.98
    aved
    0.97
    Activations Density 0.044%

    No Known Activations