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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 40092
    Prev
    Next
    INDEX
    Explanations

    terms related to illegal activities

    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
    35
    +0.14
    0.7%
    746
    +0.11
    0.5%
    67
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.14
    0.02
    35
    +0.11
    0.02
    67
    +0.11
    0.03
    Negative Logits
    efully
    -0.79
    pread
    -0.74
    ĸļ
    -0.73
    oleon
    -0.73
    Ħ¢
    -0.71
    igating
    -0.71
     Remastered
    -0.69
    ritis
    -0.68
    xual
    -0.68
    etics
    -0.67
    POSITIVE LOGITS
     immigrants
    1.01
     immigrant
    0.98
    ities
    0.98
     substances
    0.97
     alien
    0.92
     aliens
    0.92
     downloading
    0.92
     trafficking
    0.89
     dumping
    0.86
     annexation
    0.83
    Activations Density 0.022%

    No Known Activations