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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 23662
    Prev
    Next
    INDEX
    Explanations

    references to financial technology and startups

    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
    447
    +0.23
    1.1%
    481
    +0.23
    1.1%
    59
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    579
    +0.23
    0.06
    343
    +0.23
    0.05
    585
    +0.10
    0.07
    Negative Logits
     circumst
    -0.71
     canon
    -0.71
    -------------
    -0.69
     period
    -0.68
     grandchildren
    -0.67
    -------
    -0.66
    %%%%
    -0.66
     stereotypical
    -0.65
     undes
    -0.65
     nobles
    -0.65
    POSITIVE LOGITS
     Technologies
    1.07
    ®
    1.03
     Labs
    1.01
     Ltd
    1.01
    Desk
    1.01
    Wallet
    1.00
     Solutions
    0.97
    Safe
    0.97
     Analytics
    0.97
    Sense
    0.97
    Activations Density 0.607%

    No Known Activations