© 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 768
    9. 8-RES_FS768-JB
    11. 758
    Prev
    Next
    INDEX
    Explanations

    terms related to social, political, and legal issues, as well as technological terms like software and hardware

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    79
    +0.12
    0.5%
    762
    +0.11
    0.5%
    504
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.19
    762
    +0.11
    0.17
    225
    +0.11
    0.14
    Negative Logits
    pload
    -0.67
    imum
    -0.64
    Joy
    -0.62
    ËĪ
    -0.61
    atus
    -0.60
    aed
    -0.60
    iom
    -0.59
    lopp
    -0.59
    antha
    -0.59
    jan
    -0.58
    POSITIVE LOGITS
     alike
    1.93
     respectively
    1.55
     depending
    1.01
    depending
    0.98
     sectors
    0.87
     thereof
    0.86
     industries
    0.80
     versa
    0.79
    SPONSORED
    0.78
     combo
    0.68
    Activations Density 5.157%

    No Known Activations