© 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. 37299
    Prev
    Next
    INDEX
    Explanations

    references to the reader or audience in a conversational context

    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
    497
    +0.14
    0.7%
    49
    +0.13
    0.6%
    566
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    49
    +0.14
    0.07
    354
    +0.13
    0.06
    502
    +0.10
    0.04
    Negative Logits
     Excellence
    -0.69
    votes
    -0.64
     airs
    -0.63
     Innocent
    -0.62
    ween
    -0.61
    adium
    -0.61
     Political
    -0.60
    icy
    -0.59
    ensable
    -0.59
     Scarborough
    -0.57
    POSITIVE LOGITS
    're
    1.43
    'll
    1.36
    've
    1.21
     guessed
    1.20
     can
    1.06
    'd
    1.01
     guys
    0.99
     know
    0.99
     mileage
    0.99
    tub
    0.98
    Activations Density 0.120%

    No Known Activations