© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 3000
    Prev
    Next
    INDEX
    Explanations

    references to financial terms and each instance corresponds to different aspects like voting, brain-computer interfaces, money management, and legal procedures surrounding patents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    447
    +0.17
    0.8%
    481
    +0.16
    0.7%
    241
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.17
    0.17
    288
    +0.16
    0.18
    225
    +0.13
    0.14
    Negative Logits
    stood
    -0.79
     pestic
    -0.75
     GN
    -0.64
    ailability
    -0.63
    cluded
    -0.59
    audi
    -0.57
    idth
    -0.56
     willingness
    -0.55
     hypothal
    -0.54
    ofi
    -0.54
    POSITIVE LOGITS
     cheaply
    1.17
     anonymously
    1.13
     WITHOUT
    1.13
     efficiently
    1.09
     without
    1.09
     responsibly
    1.06
     abroad
    1.04
     whilst
    1.02
     peacefully
    1.01
     safely
    1.00
    Activations Density 2.534%

    No Known Activations