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. 43246
    Prev
    Next
    INDEX
    Explanations

    instances of positive sentiment or favorable evaluations

    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
    314
    +0.13
    0.6%
    137
    +0.11
    0.5%
    695
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    137
    +0.13
    0.01
    222
    +0.11
    0.01
    220
    +0.10
    0.01
    Negative Logits
    ãĤ©
    -0.84
    prise
    -0.79
    lar
    -0.76
    ãĥĪ
    -0.73
    anium
    -0.72
    mington
    -0.69
     Goff
    -0.68
    lon
    -0.68
    ning
    -0.68
    xton
    -0.68
    POSITIVE LOGITS
    /-
    1.11
     IMAGES
    1.08
    /+
    0.68
    #$
    0.67
    %.
    0.66
     intersections
    0.65
    ileged
    0.64
     crossings
    0.63
     editions
    0.61
    TIT
    0.61
    Activations Density 0.006%

    No Known Activations