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. 227
    Prev
    Next
    INDEX
    Explanations

    phrases related to scientific research and studies, especially in the field of aging and medical breakthroughs

    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
    608
    +0.20
    0.9%
    393
    +0.10
    0.5%
    90
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.20
    0.22
    669
    +0.10
    0.11
    90
    +0.10
    0.10
    Negative Logits
    minus
    -0.86
     ILCS
    -0.86
    keys
    -0.81
    hen
    -0.78
    cles
    -0.77
    hai
    -0.75
    ennis
    -0.74
    DAQ
    -0.74
    edin
    -0.72
    falls
    -0.72
    POSITIVE LOGITS
     preserving
    1.16
     obtaining
    1.11
     acquiring
    1.04
     avoiding
    1.02
     achieving
    1.02
     maintaining
    1.01
     resolving
    1.01
     creating
    1.00
     restoring
    1.00
     establishing
    1.00
    Activations Density 2.607%

    No Known Activations