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. 26337
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "du" indicating a connection to duality or partnership themes

    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
    206
    +0.12
    0.5%
    99
    +0.10
    0.5%
    326
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    673
    +0.12
    0.01
    448
    +0.10
    0.01
    657
    +0.10
    0.01
    Negative Logits
    ãģĨ
    -0.80
    picture
    -0.71
    objects
    -0.68
    deals
    -0.66
    boards
    -0.66
    clinton
    -0.65
    ISC
    -0.64
    lessness
    -0.64
    skirts
    -0.64
    IO
    -0.63
    POSITIVE LOGITS
     jour
    1.05
    ²¾
    0.86
    cci
    0.83
    Pont
    0.82
    pload
    0.74
    quet
    0.73
    perate
    0.73
     Qué
    0.72
    pper
    0.72
     lieu
    0.71
    Activations Density 0.004%

    No Known Activations