© 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. 9756
    Prev
    Next
    INDEX
    Explanations

    mentions of sports-related actions and events

    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
    481
    +0.16
    0.7%
    325
    +0.10
    0.5%
    78
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.16
    0.05
    78
    +0.10
    0.04
    326
    +0.10
    0.04
    Negative Logits
    anamo
    -0.73
     Journals
    -0.73
     staffing
    -0.71
     Cabin
    -0.70
     organising
    -0.70
     dstg
    -0.69
     membership
    -0.69
    rongh
    -0.68
    VW
    -0.68
     Flavoring
    -0.66
    POSITIVE LOGITS
     misses
    0.98
     volley
    0.86
     touchdown
    0.85
     hurled
    0.84
    handler
    0.84
     downed
    0.83
     scored
    0.82
     thrown
    0.80
    hitting
    0.80
     slam
    0.79
    Activations Density 0.257%

    No Known Activations