© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 768
    9. 8-RES_FS768-JB
    11. 658
    Prev
    Next
    INDEX
    Explanations

    terms related to legal matters and regulations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    502
    +0.12
    0.5%
    566
    +0.11
    0.5%
    430
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.12
    0.14
    502
    +0.11
    0.13
    49
    +0.09
    0.13
    Negative Logits
    geries
    -0.69
     Schiff
    -0.64
     intervening
    -0.56
     Slim
    -0.56
    onet
    -0.56
     Slav
    -0.55
    raints
    -0.54
     Bronze
    -0.54
     Serbian
    -0.53
    hess
    -0.53
    POSITIVE LOGITS
     deems
    1.12
     deem
    1.04
     sorely
    0.79
     cherish
    0.79
     encounter
    0.78
     entails
    0.78
     dearly
    0.77
     aspire
    0.75
     despise
    0.74
     encountered
    0.73
    Activations Density 2.404%

    No Known Activations