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. 26161
    Prev
    Next
    INDEX
    Explanations

    elements related to various types of materials and their associated attributes or specifications

    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
    226
    +0.12
    0.5%
    398
    +0.11
    0.5%
    609
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.04
    575
    +0.11
    0.03
    537
    +0.09
    0.03
    Negative Logits
    instead
    -0.61
    rather
    -0.59
     Versus
    -0.56
     boo
    -0.56
    Too
    -0.56
    ":"/
    -0.55
    ularity
    -0.54
    apest
    -0.54
     faire
    -0.53
     REALLY
    -0.52
    POSITIVE LOGITS
     thereto
    0.76
    anders
    0.67
    alsa
    0.67
    ,[
    0.65
    ).[
    0.64
    .[
    0.63
    monds
    0.61
    catentry
    0.60
     and
    0.58
    berra
    0.57
    Activations Density 0.181%

    No Known Activations