© 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. 26677
    Prev
    Next
    INDEX
    Explanations

    references to "Flash" in various contexts

    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
    +0.11
    0.5%
    553
    +0.10
    0.5%
    611
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.01
    611
    +0.10
    0.01
    220
    +0.10
    0.01
    Negative Logits
    mble
    -0.89
    ufact
    -0.81
    emort
    -0.80
    xual
    -0.79
    ablishment
    -0.78
    zsche
    -0.77
    ally
    -0.76
    ãĥĥãĥĪ
    -0.76
    ality
    -0.73
     Seym
    -0.72
    POSITIVE LOGITS
    light
    0.97
    bang
    0.94
    lights
    0.89
    cards
    0.89
    trap
    0.88
    Flash
    0.84
    Zone
    0.82
    point
    0.81
    fast
    0.81
     Flash
    0.81
    Activations Density 0.006%

    No Known Activations