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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 60
    Prev
    Next
    INDEX
    Explanations

    references to video game titles and characters within the Metal Gear Solid franchise

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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%
    556
    +0.10
    0.5%
    359
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    556
    +0.13
    0.05
    359
    +0.10
    0.04
    326
    +0.10
    0.04
    Negative Logits
    uckland
    -0.80
    shr
    -0.80
    umbn
    -0.80
    rist
    -0.79
     eleph
    -0.78
    mean
    -0.76
    ask
    -0.76
    Reviewer
    -0.76
    abo
    -0.75
    elong
    -0.75
    POSITIVE LOGITS
    âĦ¢:
    1.05
     Tactics
    1.05
     III
    1.03
    ®
    1.01
     Primal
    1.00
     Origins
    0.98
     Generations
    0.96
     Legends
    0.94
     XIV
    0.94
     IV
    0.92
    Activations Density 0.225%

    No Known Activations