© 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 24576
    9. 8-RES_FS24576-JB
    11. 9383
    Prev
    Next
    INDEX
    Explanations

    interesting adjectives and descriptors

    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
    131
    +0.10
    0.4%
    404
    +0.09
    0.4%
    81
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.10
    0.03
    429
    +0.09
    0.03
    37
    +0.09
    0.03
    Negative Logits
    ÄŁ
    -0.82
    enting
    -0.75
    erning
    -0.74
    icipated
    -0.74
    essing
    -0.74
    arest
    -0.72
    arthy
    -0.72
    ighty
    -0.71
    DERR
    -0.68
    adier
    -0.68
    POSITIVE LOGITS
    NESS
    0.97
    river
    0.89
    ness
    0.81
    ges
    0.72
    rive
    0.71
    glers
    0.70
     enterprise
    0.67
    nesses
    0.66
     substance
    0.65
    paren
    0.64
    Activations Density 0.171%

    No Known Activations