Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1308
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "to," often used to indicate infinitive verbs

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    665
    +0.20
    1.0%
    608
    +0.16
    0.8%
    718
    +0.16
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.20
    0.24
    718
    +0.16
    0.21
    49
    +0.16
    0.21
    Negative Logits
    quickShipAvailable
    -0.81
    soDeliveryDate
    -0.75
    calling
    -0.68
     Closing
    -0.68
    listed
    -0.67
    rising
    -0.67
     Explan
    -0.65
    points
    -0.65
     meant
    -0.63
     Caller
    -0.63
    POSITIVE LOGITS
     pursue
    1.39
     participate
    1.37
     declare
    1.33
     engage
    1.32
     undertake
    1.32
     negotiate
    1.31
     accumulate
    1.29
     defend
    1.29
     perform
    1.29
     explore
    1.28
    Activations Density 1.684%

    No Known Activations