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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1242
    Prev
    Next
    INDEX
    Explanations

    information related to military conflicts in Syria

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    447
    +0.13
    0.6%
    481
    +0.12
    0.6%
    288
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.16
    57
    +0.12
    0.09
    612
    +0.12
    0.09
    Negative Logits
    Premium
    -0.76
    Seattle
    -0.74
    ertodd
    -0.71
     Wilde
    -0.70
    Fish
    -0.70
    Hug
    -0.69
    Consumer
    -0.69
     Phelps
    -0.68
    inburgh
    -0.68
    inarily
    -0.68
    POSITIVE LOGITS
     insurgents
    1.09
     Hussein
    1.07
     refugees
    1.05
     militias
    1.03
     refugee
    1.02
     ceasefire
    1.00
     militants
    0.97
     Arab
    0.97
     rebels
    0.96
    Assad
    0.96
    Activations Density 1.966%

    No Known Activations