Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 37190
    Prev
    Next
    INDEX
    Explanations

    mentions of specific counties and judicial matters involving police actions

    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
    715
    +0.10
    0.5%
    752
    +0.10
    0.5%
    731
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    611
    +0.10
    0.02
    667
    +0.10
    0.02
    538
    +0.10
    0.02
    Negative Logits
    advertisement
    -0.89
     facult
    -0.80
    asy
    -0.75
     streng
    -0.74
     artif
    -0.74
     tink
    -0.73
    chairs
    -0.72
    isSpecialOrderable
    -0.71
     soap
    -0.70
     embr
    -0.69
    POSITIVE LOGITS
     County
    1.97
     COUNTY
    1.48
     county
    1.22
     Valley
    1.18
     counties
    1.18
     Township
    1.12
     Parish
    1.07
     Circuit
    1.07
     Islands
    1.05
    VILLE
    1.05
    Activations Density 0.044%

    No Known Activations