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. 23307
    Prev
    Next
    INDEX
    Explanations

    references to the working class and its economic conditions

    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
    288
    +0.14
    0.7%
    447
    +0.10
    0.5%
    152
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    18
    +0.14
    0.01
    288
    +0.10
    0.01
    152
    +0.10
    0.01
    Negative Logits
    inventoryQuantity
    -0.96
    WAYS
    -0.83
    oldown
    -0.81
    BIT
    -0.75
    Pokemon
    -0.75
    Ħ¢
    -0.74
    orius
    -0.73
    forge
    -0.71
    ococ
    -0.71
    etsk
    -0.70
    POSITIVE LOGITS
     revolt
    0.88
     revolutionaries
    0.82
     peasants
    0.82
     solidarity
    0.81
     mobility
    0.77
     mobilization
    0.77
     peasant
    0.77
     uprising
    0.77
     movement
    0.74
     empowerment
    0.74
    Activations Density 0.012%

    No Known Activations