© 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. 7925
    Prev
    Next
    INDEX
    Explanations

    words related to scientific or technical documentation

    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
    461
    +0.11
    0.5%
    276
    +0.09
    0.4%
    568
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    568
    +0.11
    0.02
    28
    +0.09
    0.02
    461
    +0.09
    0.02
    Negative Logits
     Lauder
    -0.73
    ILLE
    -0.70
    IGHTS
    -0.68
    BIP
    -0.66
    taboola
    -0.65
    WAYS
    -0.60
     Sins
    -0.60
    è£
    -0.59
    CRIPTION
    -0.59
     FANT
    -0.59
    POSITIVE LOGITS
    inion
    1.40
    yright
    1.29
    olitan
    1.25
    rompt
    1.21
    edia
    1.18
    sych
    1.15
    odcast
    1.09
    arent
    1.07
    ocalypse
    1.05
    lication
    1.05
    Activations Density 0.025%

    No Known Activations