© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 45739
    Prev
    Next
    INDEX
    Explanations

    forms of the verb "to be" in various tenses

    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
    55
    +0.16
    0.7%
    102
    +0.12
    0.5%
    528
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.16
    0.05
    102
    +0.12
    0.03
    458
    +0.10
    0.03
    Negative Logits
    bilt
    -0.72
    itches
    -0.69
    erald
    -0.64
    oso
    -0.61
    itures
    -0.61
    Footnote
    -0.60
    pox
    -0.59
    stice
    -0.59
     Moines
    -0.59
    vale
    -0.59
    POSITIVE LOGITS
     happening
    1.36
     occurring
    0.97
     going
    0.93
     happ
    0.91
     transpired
    0.90
     meant
    0.89
     unfolding
    0.88
     done
    0.87
     wrong
    0.85
     gonna
    0.84
    Activations Density 0.055%

    No Known Activations