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 49152
    5. 8-RES_FS49152-JB
    6. 43167
    Prev
    Next
    INDEX
    Explanations

    references to environmental issues and warnings about the consequences of overexploitation

    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
    640
    +0.27
    1.3%
    442
    +0.16
    0.8%
    160
    +0.15
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.27
    0.12
    442
    +0.16
    0.11
    160
    +0.15
    0.10
    Negative Logits
     skelet
    -0.71
     sculpt
    -0.70
     vict
    -0.68
     champion
    -0.67
     Saiyan
    -0.66
     occult
    -0.65
     Cind
    -0.65
     honors
    -0.65
    ©¶æ
    -0.65
     waterfall
    -0.65
    POSITIVE LOGITS
    However
    1.60
    Refer
    1.51
    Such
    1.48
    Asked
    1.48
    Instead
    1.48
    Besides
    1.47
    SPONSORED
    1.46
    Calling
    1.46
    Similarly
    1.45
    But
    1.45
    Activations Density 0.274%

    No Known Activations