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. 43213
    Prev
    Next
    INDEX
    Explanations

    financial obligations and payments

    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
    556
    +0.10
    0.5%
    25
    +0.09
    0.4%
    113
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    552
    +0.10
    0.03
    604
    +0.09
    0.03
    25
    +0.09
    0.03
    Negative Logits
    bum
    -0.66
    bably
    -0.63
    GGGGGGGG
    -0.62
    ãĥIJ
    -0.61
    ĸļ
    -0.61
    aughlin
    -0.60
     guiActiveUnfocused
    -0.59
     Moder
    -0.59
    è¯
    -0.59
    \/\/
    -0.59
    POSITIVE LOGITS
     dues
    1.24
     taxes
    1.20
     debts
    1.13
     Taxes
    1.08
     royalties
    1.07
     attention
    1.03
     dividends
    1.00
     fees
    0.99
     owed
    0.98
     bills
    0.97
    Activations Density 0.028%

    No Known Activations