© 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. 4440
    Prev
    Next
    INDEX
    Explanations

    modal verbs and phrases indicating possibility or potential outcomes

    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
    373
    +0.14
    0.6%
    569
    +0.10
    0.5%
    737
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.14
    0.05
    718
    +0.10
    0.04
    139
    +0.10
    0.03
    Negative Logits
     wasting
    -0.71
     accountable
    -0.71
     wary
    -0.67
     ineligible
    -0.66
     plotting
    -0.66
     foregoing
    -0.65
     proficient
    -0.65
     glad
    -0.64
     polite
    -0.63
     thankful
    -0.63
    POSITIVE LOGITS
     reside
    0.88
    PsyNetMessage
    0.86
    KEN
    0.82
     involve
    0.78
    ãĤ«
    0.76
    âĦ¢:
    0.76
     hinges
    0.73
    xxxxxxxx
    0.72
    :{
    0.71
     arise
    0.70
    Activations Density 0.204%

    No Known Activations