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 1536
    5. 8-RES_FS1536-JB
    6. 1481
    Prev
    Next
    INDEX
    Explanations

    words related to time and continuous action

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    218
    +0.16
    0.8%
    450
    +0.10
    0.5%
    590
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    218
    +0.16
    0.06
    98
    +0.10
    0.06
    306
    +0.09
    0.06
    Negative Logits
    pour
    -0.78
    puff
    -0.66
    Desc
    -0.66
    eny
    -0.65
    eri
    -0.65
    esta
    -0.64
    VP
    -0.64
    VG
    -0.64
     Diary
    -0.64
    cart
    -0.63
    POSITIVE LOGITS
     proven
    0.87
     adulthood
    0.78
     AFTER
    0.74
    soever
    0.74
     someone
    0.71
     reinforcements
    0.70
     2024
    0.70
     they
    0.70
     sunrise
    0.69
     somebody
    0.68
    Activations Density 1.125%

    No Known Activations