© 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 49152
    9. 8-RES_FS49152-JB
    11. 37353
    Prev
    Next
    INDEX
    Explanations

    references to political candidates

    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
    +0.12
    0.5%
    371
    +0.11
    0.5%
    72
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.12
    0.02
    224
    +0.11
    0.02
    371
    +0.11
    0.02
    Negative Logits
    erie
    -0.81
    ãĥ¼ãĥĨãĤ£
    -0.69
    orgetown
    -0.67
    ^^^^
    -0.67
    cest
    -0.67
    OME
    -0.67
    iencies
    -0.66
    ories
    -0.64
    EMA
    -0.63
    ãĥĥ
    -0.63
    POSITIVE LOGITS
    idates
    0.92
     candidate
    0.91
     candidates
    0.90
    cand
    0.81
     Candidate
    0.78
     voters
    0.75
    ndum
    0.73
     frontrunner
    0.73
     ballots
    0.72
     Narendra
    0.71
    Activations Density 0.023%

    No Known Activations