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 24576
    5. 8-RES_FS24576-JB
    6. 19109
    Prev
    Next
    INDEX
    Explanations

    positive expressions or remarks

    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
    +0.17
    0.9%
    745
    +0.10
    0.5%
    534
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    411
    +0.17
    0.04
    718
    +0.10
    0.04
    41
    +0.09
    0.04
    Negative Logits
     periodically
    -0.65
    pour
    -0.55
    bil
    -0.55
    lia
    -0.53
     increment
    -0.52
     periodic
    -0.52
     corpus
    -0.51
     utility
    -0.51
     exit
    -0.50
     backdoor
    -0.50
    POSITIVE LOGITS
    efe
    0.72
    quite
    0.63
     dreamed
    0.60
     matched
    0.59
    ented
    0.59
    quickShipAvailable
    0.57
    ħĭ
    0.57
     afforded
    0.56
     hotter
    0.56
    bleacher
    0.55
    Activations Density 0.209%

    No Known Activations