© 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 EvalsExportsinterp-engineNEWAPI 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. 2633
    Prev
    Next
    INDEX
    Explanations

    phrases related to sharing opinions or sentiments publicly

    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
    380
    +0.12
    0.6%
    195
    +0.12
    0.5%
    574
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    380
    +0.12
    0.03
    195
    +0.12
    0.02
    242
    +0.08
    0.02
    Negative Logits
    ctors
    -0.67
    ade
    -0.66
    osate
    -0.65
     Journals
    -0.65
    é¾įå
    -0.63
     Straight
    -0.63
    ney
    -0.63
    lez
    -0.62
     Disciple
    -0.62
     Journalists
    -0.61
    POSITIVE LOGITS
     displeasure
    1.37
     remorse
    1.15
     gratitude
    1.15
     condolences
    1.14
     frustration
    1.13
     frustrations
    1.12
     doubts
    1.11
     willingness
    1.09
     concerns
    1.08
     sympathy
    1.06
    Activations Density 0.069%

    No Known Activations