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 49152
    5. 8-RES_FS49152-JB
    6. 46044
    Prev
    Next
    INDEX
    Explanations

    phrases and contexts related to aspirations and ambitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    641
    +0.11
    0.5%
    288
    +0.10
    0.5%
    575
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    641
    +0.11
    0.02
    575
    +0.10
    0.02
    247
    +0.10
    0.02
    Negative Logits
    avis
    -0.78
    arius
    -0.73
    ãĥĺãĥ©
    -0.72
    idges
    -0.68
    ensibly
    -0.66
    Fed
    -0.66
    anti
    -0.65
    ols
    -0.65
    Interstitial
    -0.64
    atin
    -0.63
    POSITIVE LOGITS
    tek
    0.90
    scape
    0.85
    hack
    0.84
     dream
    0.81
    liner
    0.81
     fulfilled
    0.81
     dreaming
    0.79
     dreams
    0.77
    dream
    0.77
    bler
    0.76
    Activations Density 0.014%

    No Known Activations