© 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. 7292
    Prev
    Next
    INDEX
    Explanations

    mentions of specific people and their roles in various events or situations

    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
    373
    +0.17
    0.8%
    426
    +0.12
    0.6%
    591
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.17
    0.04
    430
    +0.12
    0.03
    426
    +0.10
    0.02
    Negative Logits
     derog
    -0.73
     signalling
    -0.72
    withstanding
    -0.71
     discontin
    -0.70
    HAHA
    -0.70
    anian
    -0.69
     maiden
    -0.66
    ogyn
    -0.64
     marginal
    -0.63
     blacklist
    -0.63
    POSITIVE LOGITS
     VIDEOS
    0.94
     Caption
    0.93
     Plus
    0.90
    PHOTOS
    0.87
     Courtesy
    0.86
    Photos
    0.83
     Replay
    0.83
     POLITICO
    0.82
     Hear
    0.82
     Learn
    0.81
    Activations Density 0.135%

    No Known Activations