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 768
    5. 8-RES_FS768-JB
    6. 543
    Prev
    Next
    INDEX
    Explanations

    terms related to naval activities and events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    578
    +0.15
    0.7%
    274
    +0.11
    0.5%
    658
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.15
    0.14
    659
    +0.11
    0.11
    346
    +0.10
    0.10
    Negative Logits
    anwhile
    -0.70
     numbered
    -0.69
    etheless
    -0.65
     Medieval
    -0.59
    åŃ
    -0.59
     adren
    -0.58
     Mayhem
    -0.58
    CAST
    -0.57
    ocene
    -0.57
    onymous
    -0.56
    POSITIVE LOGITS
    IELD
    1.23
    adows
    1.07
    ooting
    1.04
    allow
    1.03
    uffle
    1.01
    apesh
    1.00
    aping
    1.00
    apeshifter
    0.99
    apes
    0.97
    bour
    0.97
    Activations Density 3.271%

    No Known Activations