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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18762
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to politics and environmental organizations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.27
    1.3%
    481
    +0.23
    1.1%
    186
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    461
    +0.27
    0.06
    401
    +0.23
    0.07
    481
    +0.12
    0.08
    Negative Logits
    interstitial
    -0.89
    taboola
    -0.75
    ],"
    -0.69
    groupon
    -0.64
     guiActiveUnfocused
    -0.62
    ãĥª
    -0.60
    ]+
    -0.59
    upper
    -0.58
    "],
    -0.58
    externalActionCode
    -0.58
    POSITIVE LOGITS
     therefore
    1.03
     however
    1.01
     meanwhile
    1.00
     certainly
    0.98
     wisely
    0.91
     solves
    0.89
     notwithstanding
    0.85
     famously
    0.83
     assures
    0.82
     understandably
    0.82
    Activations Density 0.632%

    No Known Activations