© 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 49152
    9. 8-RES_FS49152-JB
    11. 41613
    Prev
    Next
    INDEX
    Explanations

    words and phrases expressing positive emotions or excitement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    407
    +0.12
    0.5%
    571
    +0.12
    0.5%
    201
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.12
    0.02
    571
    +0.12
    0.02
    317
    +0.10
    0.02
    Negative Logits
    rame
    -0.77
     landfill
    -0.70
    arij
    -0.69
     flawed
    -0.68
     faulty
    -0.67
    lder
    -0.67
     detrimental
    -0.64
     restrictive
    -0.64
    issors
    -0.63
     restraining
    -0.62
    POSITIVE LOGITS
    iously
    0.99
     delight
    0.93
     ecstatic
    0.90
     delighted
    0.87
     anticipation
    0.86
     exclaim
    0.85
     grin
    0.85
     gle
    0.83
     eagerly
    0.81
     praise
    0.80
    Activations Density 0.049%

    No Known Activations