© 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 12288
    9. 8-RES_FS12288-JB
    11. 9364
    Prev
    Next
    INDEX
    Explanations

    company or product names

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    250
    +0.11
    0.5%
    677
    +0.10
    0.5%
    758
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.05
    392
    +0.10
    0.05
    659
    +0.10
    0.04
    Negative Logits
    inventoryQuantity
    -0.82
    soType
    -0.81
     Contra
    -0.78
     defic
    -0.76
     antic
    -0.74
    ENE
    -0.74
     antit
    -0.69
     dilig
    -0.68
     confir
    -0.68
     antibiotic
    -0.67
    POSITIVE LOGITS
    k
    1.37
    ks
    1.30
    kt
    1.28
    K
    1.26
    ked
    1.25
    king
    1.20
    KE
    1.18
    ke
    1.18
    kb
    1.18
    KC
    1.17
    Activations Density 0.274%

    No Known Activations