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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 43375
    Prev
    Next
    INDEX
    Explanations

    the verb "be" in various forms and contexts

    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
    102
    +0.16
    0.7%
    566
    +0.13
    0.6%
    55
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.16
    0.03
    566
    +0.13
    0.03
    715
    +0.10
    0.03
    Negative Logits
    ongyang
    -0.72
     rigs
    -0.69
    plings
    -0.65
    ortmund
    -0.64
     strain
    -0.64
    aceutical
    -0.62
    rones
    -0.62
    strous
    -0.62
     dispute
    -0.62
     Bever
    -0.61
    POSITIVE LOGITS
     able
    1.22
     reunited
    0.91
     reminded
    0.86
     reborn
    0.83
     surrounded
    0.83
     challenged
    0.83
     treated
    0.82
     bitten
    0.82
     seated
    0.82
     aware
    0.81
    Activations Density 0.040%

    No Known Activations