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 24576
    5. 8-RES_FS24576-JB
    6. 21697
    Prev
    Next
    INDEX
    Explanations

    emotions or attitudes such as hope, fear, anxiety, fear, reverence, conviction, optimism, scepticism, cynicism, remorse, outrage in a context

    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
    481
    +0.14
    0.6%
    760
    +0.10
    0.4%
    170
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    384
    +0.14
    0.04
    393
    +0.10
    0.05
    595
    +0.10
    0.03
    Negative Logits
    kees
    -0.93
    »Ĵ
    -0.91
    ogun
    -0.85
    enium
    -0.84
     Cosponsors
    -0.83
    ucci
    -0.81
    married
    -0.79
     Nare
    -0.79
    vae
    -0.77
    sites
    -0.77
    POSITIVE LOGITS
     indifference
    1.26
     cynicism
    1.26
     complicity
    1.24
     frustration
    1.19
     hostility
    1.18
     humility
    1.17
     emotion
    1.17
     ignorance
    1.15
     sincerity
    1.15
     passion
    1.15
    Activations Density 0.301%

    No Known Activations