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

    Neuronpedia

    Jacobian LensNEW
    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. 7633
    Prev
    Next
    INDEX
    Explanations

    requests to answer basic questions as part of an application or survey process

    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
    355
    +0.11
    0.5%
    245
    +0.10
    0.5%
    734
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    163
    +0.11
    0.02
    250
    +0.10
    0.01
    355
    +0.10
    0.02
    Negative Logits
     Signed
    -0.65
    Cath
    -0.65
    ONSORED
    -0.62
    rawdownloadcloneembedreportprint
    -0.60
    fitted
    -0.60
     survived
    -0.59
    Tools
    -0.59
    weights
    -0.59
     compensated
    -0.59
    bourg
    -0.58
    POSITIVE LOGITS
     questions
    1.30
     question
    1.26
     queries
    1.12
     query
    1.03
     questionnaire
    0.98
    question
    0.97
     QUEST
    0.96
     Questions
    0.95
     Question
    0.94
     unanswered
    0.94
    Activations Density 0.108%

    No Known Activations