© 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. 6985
    Prev
    Next
    INDEX
    Explanations

    references to legal terms, historical documents, and locations

    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
    238
    +0.13
    0.6%
    167
    +0.13
    0.6%
    59
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.13
    0.02
    232
    +0.13
    0.02
    251
    +0.11
    0.02
    Negative Logits
    lihood
    -0.93
    CRIP
    -0.83
    INESS
    -0.73
    TPS
    -0.73
    terday
    -0.71
    éĹĺ
    -0.69
     Dangerous
    -0.69
    UGE
    -0.66
    Magikarp
    -0.63
    ï¸
    -0.62
    POSITIVE LOGITS
    fman
    1.14
    ruption
    1.12
    rupt
    1.10
    rupted
    1.09
    relation
    1.09
    bett
    1.06
    nered
    1.06
    porate
    1.06
    bis
    1.04
    byn
    1.02
    Activations Density 0.014%

    No Known Activations