© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 519
    Prev
    Next
    INDEX
    Explanations

    terms related to finance and finance ministers

    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
    288
    +0.15
    0.7%
    635
    +0.14
    0.6%
    195
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.02
    132
    +0.14
    0.02
    286
    +0.12
    0.02
    Negative Logits
    iple
    -0.74
    76561
    -0.73
    ifiers
    -0.71
    inel
    -0.67
    Magikarp
    -0.66
    imus
    -0.66
    imum
    -0.66
    gom
    -0.64
    imens
    -0.64
    agonist
    -0.63
    POSITIVE LOGITS
    etta
    0.84
    committee
    0.79
    woman
    0.78
    notes
    0.78
    ably
    0.78
    ament
    0.78
     transactions
    0.75
    rollers
    0.75
     laundering
    0.73
    ements
    0.72
    Activations Density 0.056%

    No Known Activations