© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 41477
    Prev
    Next
    INDEX
    Explanations

    the modal verb "would" indicating hypothetical or conditional situations

    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
    339
    +0.10
    0.5%
    305
    +0.10
    0.5%
    551
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.10
    0.06
    139
    +0.10
    0.05
    661
    +0.10
    0.04
    Negative Logits
    è¦ļéĨĴ
    -0.80
     Nieto
    -0.69
     Corpus
    -0.66
     Jav
    -0.63
     Shopping
    -0.62
     Balanced
    -0.61
    strings
    -0.60
     Indust
    -0.60
     Vis
    -0.60
     Chal
    -0.59
    POSITIVE LOGITS
     doubtless
    1.05
     dearly
    0.96
     be
    0.94
     suffice
    0.91
     undoubtedly
    0.89
     imply
    0.88
     characterize
    0.87
     surely
    0.85
     later
    0.85
     ordinarily
    0.84
    Activations Density 0.112%

    No Known Activations