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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 768
    5. 8-RES_FS768-JB
    6. 327
    Prev
    Next
    INDEX
    Explanations

    religious terms and concepts

    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
    7
    +0.11
    0.5%
    477
    +0.10
    0.5%
    103
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    560
    +0.11
    0.05
    737
    +0.10
    0.07
    477
    +0.10
    0.04
    Negative Logits
     Trend
    -0.92
     GoPro
    -0.92
     Spotlight
    -0.92
     trending
    -0.87
     sourcing
    -0.86
     CPC
    -0.85
     Deadline
    -0.85
     takeaway
    -0.84
     staffers
    -0.83
     booking
    -0.81
    POSITIVE LOGITS
    ,—
    1.31
     consec
    1.31
     thence
    1.30
     hereafter
    1.30
     divine
    1.29
     everlasting
    1.24
     hath
    1.21
     whence
    1.20
     thou
    1.19
     heavenly
    1.16
    Activations Density 6.931%

    No Known Activations