© 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 12288
    9. 8-RES_FS12288-JB
    11. 10845
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for specific names, possibly related to locations or organizations

    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
    +0.18
    0.9%
    447
    +0.12
    0.6%
    186
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.18
    0.07
    146
    +0.12
    0.08
    401
    +0.11
    0.08
    Negative Logits
    piring
    -0.65
     streak
    -0.62
    romeda
    -0.61
     Adin
    -0.61
     liking
    -0.60
     revolving
    -0.59
    range
    -0.58
    )",
    -0.54
    spread
    -0.53
    interstitial
    -0.53
    POSITIVE LOGITS
     Ltd
    1.11
     employs
    1.08
     operates
    1.06
     expects
    1.05
     believes
    1.01
     maintains
    1.01
     intends
    1.00
     understands
    0.99
     has
    0.99
     publishes
    0.99
    Activations Density 0.731%

    No Known Activations