© 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. 9870
    Prev
    Next
    INDEX
    Explanations

    mentions of the name ''Ren'' or variations of it within a context related to politics, locations, or legal matters

    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
    223
    +0.11
    0.5%
    114
    +0.10
    0.4%
    339
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.02
    326
    +0.10
    0.02
    746
    +0.10
    0.01
    Negative Logits
    stakes
    -0.65
     NETWORK
    -0.64
     Antar
    -0.62
     Beir
    -0.61
    Ö¼
    -0.60
     Hemp
    -0.60
     Mayhem
    -0.60
     incidentally
    -0.59
     mileage
    -0.59
     Continental
    -0.58
    POSITIVE LOGITS
    unciation
    1.22
    unci
    1.15
    owned
    1.11
    issance
    1.07
    ovation
    1.07
    emy
    1.05
    ault
    1.04
    oir
    1.02
    ouncing
    1.00
    umbered
    1.00
    Activations Density 0.076%

    No Known Activations