Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 37096
    Prev
    Next
    INDEX
    Explanations

    punctuated quotes

    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
    102
    +0.18
    0.8%
    160
    +0.13
    0.6%
    314
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.18
    0.05
    160
    +0.13
    0.04
    442
    +0.13
    0.04
    Negative Logits
     accomp
    -0.76
     arri
    -0.76
     derby
    -0.71
     roundup
    -0.71
     metic
    -0.69
     roar
    -0.68
     nasal
    -0.68
     rout
    -0.68
     eviction
    -0.68
     redes
    -0.67
    POSITIVE LOGITS
    there
    1.64
    many
    1.45
    nothing
    1.43
    every
    1.43
    someone
    1.41
    everything
    1.40
    everyone
    1.39
    they
    1.39
    nob
    1.38
    we
    1.38
    Activations Density 0.056%

    No Known Activations