© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 37414
    Prev
    Next
    INDEX
    Explanations

    terms related to game features and management functions

    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%
    281
    +0.13
    0.6%
    415
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.13
    0.06
    459
    +0.13
    0.05
    288
    +0.10
    0.06
    Negative Logits
    inventoryQuantity
    -1.00
    姫
    -0.89
    ãģĦ
    -0.76
    tics
    -0.71
    ãĥ©ãĥ³
    -0.70
     Reps
    -0.68
    ãģĹ
    -0.66
     Cosponsors
    -0.65
    ths
    -0.65
    apeake
    -0.65
    POSITIVE LOGITS
     consisting
    1.23
     featuring
    1.01
     comprised
    0.90
     showcasing
    0.86
     comprising
    0.84
     dedicated
    0.84
     staffed
    0.84
     headlined
    0.81
     resembling
    0.81
     wherein
    0.80
    Activations Density 0.280%

    No Known Activations