© 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 24576
    9. 8-RES_FS24576-JB
    11. 9394
    Prev
    Next
    INDEX
    Explanations

    names of companies and places associated with construction and contracts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    194
    +0.11
    0.5%
    462
    +0.09
    0.4%
    659
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.06
    145
    +0.09
    0.05
    343
    +0.09
    0.05
    Negative Logits
    "))
    -0.76
     Practices
    -0.56
     fundamentally
    -0.56
    resy
    -0.56
    CONCLUS
    -0.55
    SPONSORED
    -0.55
    METHOD
    -0.55
    Conclusion
    -0.54
     disadvantages
    -0.54
    OULD
    -0.54
    POSITIVE LOGITS
    ®,
    0.78
    *,
    0.69
     ($
    0.65
    ,[
    0.63
     (#
    0.63
    !,
    0.62
     Aval
    0.61
     and
    0.59
    ,
    0.58
    Cong
    0.56
    Activations Density 0.606%

    No Known Activations