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. 23341
    Prev
    Next
    INDEX
    Explanations

    expressions of personal experience and emotional responses

    oai_token-act-pair · gpt-4o-miniTriggered 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
    480
    +0.32
    1.6%
    373
    +0.15
    0.7%
    351
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    267
    +0.32
    0.07
    598
    +0.15
    0.04
    351
    +0.10
    0.04
    Negative Logits
    hovah
    -0.73
    azeera
    -0.70
    èĢħ
    -0.70
    engers
    -0.69
    negie
    -0.66
     enshr
    -0.66
     sovereign
    -0.65
     nation
    -0.64
    enance
    -0.63
     constituency
    -0.62
    POSITIVE LOGITS
     luckily
    1.11
     Turns
    1.07
     haha
    0.92
    didn
    0.92
    Luckily
    0.86
     fortunately
    0.83
    RESULTS
    0.81
     Luckily
    0.81
     hindsight
    0.81
     I
    0.80
    Activations Density 0.720%

    No Known Activations