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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 2906
    Prev
    Next
    INDEX
    Explanations

    emotional states related to personal experiences

    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
    407
    +0.11
    0.5%
    394
    +0.10
    0.5%
    185
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.11
    0.13
    351
    +0.10
    0.06
    718
    +0.10
    0.10
    Negative Logits
    arthed
    -0.90
    senal
    -0.85
    venue
    -0.79
     Manufacturing
    -0.74
    specified
    -0.74
     patents
    -0.72
     USDA
    -0.70
    winner
    -0.70
    ospons
    -0.69
     Patent
    -0.68
    POSITIVE LOGITS
     sadness
    1.46
     sorrow
    1.37
     feeling
    1.35
     numb
    1.33
     guilt
    1.30
     longing
    1.30
     anguish
    1.27
     wondering
    1.27
     despair
    1.25
     tears
    1.25
    Activations Density 3.269%

    No Known Activations