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. 41351
    Prev
    Next
    INDEX
    Explanations

    mentions of the city of Paris

    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
    7
    +0.11
    0.5%
    481
    +0.11
    0.5%
    36
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.11
    0.02
    511
    +0.11
    0.02
    145
    +0.10
    0.02
    Negative Logits
    uilt
    -0.83
    ITH
    -0.79
    estern
    -0.74
    arijuana
    -0.69
    pta
    -0.68
    rha
    -0.68
    ocument
    -0.67
    ramid
    -0.67
    isSpecialOrderable
    -0.67
    atcher
    -0.66
    POSITIVE LOGITS
     Hilton
    1.19
    ienne
    1.03
    ian
    1.00
    ians
    0.94
    furt
    0.88
     Mé
    0.86
     Attacks
    0.84
    iens
    0.82
    bourg
    0.80
    etta
    0.79
    Activations Density 0.007%

    No Known Activations