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. 43267
    Prev
    Next
    INDEX
    Explanations

    specific combinations of Unicode characters or symbols

    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
    420
    +0.13
    0.6%
    314
    +0.12
    0.6%
    48
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    141
    +0.13
    0.00
    284
    +0.12
    0.00
    699
    +0.10
    0.00
    Negative Logits
     gripping
    -0.70
    oted
    -0.69
     caregivers
    -0.69
     skelet
    -0.66
     Hawk
    -0.64
     appre
    -0.63
    izations
    -0.61
    ousel
    -0.61
    UNCH
    -0.60
     disemb
    -0.60
    POSITIVE LOGITS
    ¿
    0.95
    enger
    0.94
    hof
    0.93
    ername
    0.93
    wald
    0.92
    Ĩ
    0.89
    gang
    0.88
    sf
    0.88
    rss
    0.87
    ĩ
    0.85
    Activations Density 0.002%

    No Known Activations