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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 11358
    Prev
    Next
    INDEX
    Explanations

    dates mentioned in the text

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.19
    0.9%
    306
    +0.11
    0.5%
    55
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.19
    0.04
    546
    +0.11
    0.03
    433
    +0.10
    0.03
    Negative Logits
    Administ
    -0.61
     FTP
    -0.60
    natureconservancy
    -0.58
     surn
    -0.57
     scene
    -0.57
     paternal
    -0.56
     interaction
    -0.56
    rw
    -0.55
     footprint
    -0.55
    itaire
    -0.54
    POSITIVE LOGITS
     2017
    1.08
     2014
    1.08
     2015
    1.08
     2019
    1.06
     2016
    1.06
     2018
    1.05
     2013
    0.98
     2001
    0.95
     2012
    0.95
     2011
    0.92
    Activations Density 0.108%

    No Known Activations