Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 45680
    Prev
    Next
    INDEX
    Explanations

    references to years and numerical data points related to demographic or economic trends

    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
    430
    +0.12
    0.5%
    338
    +0.11
    0.5%
    715
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.12
    0.02
    322
    +0.11
    0.02
    552
    +0.10
    0.02
    Negative Logits
    ensical
    -0.71
    roid
    -0.70
    rat
    -0.61
     rag
    -0.61
     pri
    -0.61
    Ax
    -0.60
    Rat
    -0.60
     corrid
    -0.60
     dissatisf
    -0.59
     rug
    -0.58
    POSITIVE LOGITS
     2010
    1.48
     2008
    1.45
     2009
    1.43
     2002
    1.39
     2007
    1.36
     1998
    1.35
     1999
    1.35
     2012
    1.34
     2005
    1.33
     2003
    1.32
    Activations Density 0.074%

    No Known Activations