© 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. 26601
    Prev
    Next
    INDEX
    Explanations

    references to online multiplayer role-playing games (MMORPGs)

    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
    459
    +0.13
    0.6%
    288
    +0.12
    0.5%
    649
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.01
    444
    +0.12
    0.01
    614
    +0.12
    0.01
    Negative Logits
    ãĥ´ãĤ¡
    -0.84
     McCartney
    -0.69
    hold
    -0.68
    patrick
    -0.68
    amus
    -0.66
     toile
    -0.64
     Coul
    -0.63
    ulkan
    -0.63
    lift
    -0.62
    ku
    -0.62
    POSITIVE LOGITS
    ORPG
    1.99
    QB
    1.19
    FG
    1.02
    TP
    0.98
    MM
    0.94
    ONEY
    0.91
    ITH
    0.89
    SE
    0.87
    DR
    0.87
    DA
    0.85
    Activations Density 0.006%

    No Known Activations