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 24576
    5. 8-RES_FS24576-JB
    6. 18976
    Prev
    Next
    INDEX
    Explanations

    information related to legal appeals and procedures

    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
    640
    +0.15
    0.7%
    217
    +0.12
    0.5%
    373
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.15
    0.06
    217
    +0.12
    0.04
    292
    +0.11
    0.03
    Negative Logits
     spilled
    -0.75
     amused
    -0.70
     McMaster
    -0.68
     Brock
    -0.67
     phony
    -0.65
     McA
    -0.65
     Pike
    -0.63
     dangling
    -0.62
    acan
    -0.62
     Dow
    -0.62
    POSITIVE LOGITS
    Well
    1.12
    Generally
    1.07
    Answer
    1.04
    Probably
    0.99
    Depending
    0.97
    Firstly
    0.96
    Absolutely
    0.96
    Due
    0.94
    Simply
    0.94
    Nothing
    0.94
    Activations Density 0.119%

    No Known Activations