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. 23747
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "was."

    oai_token-act-pair · gpt-4o-miniTriggered 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
    699
    +0.15
    0.7%
    377
    +0.11
    0.5%
    7
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.15
    0.10
    102
    +0.11
    0.09
    566
    +0.10
    0.08
    Negative Logits
    currently
    -0.75
    rouse
    -0.72
    today
    -0.68
     Current
    -0.65
    uras
    -0.64
    has
    -0.63
    ems
    -0.63
    veland
    -0.63
    oglu
    -0.62
    icles
    -0.62
    POSITIVE LOGITS
    hes
    1.08
     intentional
    1.05
     instrumental
    1.02
     originally
    0.98
     unintentional
    0.98
     intentionally
    0.97
     mistaken
    0.97
     purposely
    0.93
     careless
    0.93
     intended
    0.90
    Activations Density 0.347%

    No Known Activations