© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 37405
    Prev
    Next
    INDEX
    Explanations

    song titles or phrases related to optimism and overcoming struggles

    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
    232
    +0.14
    0.7%
    6
    +0.11
    0.5%
    430
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    431
    +0.14
    0.03
    232
    +0.11
    0.03
    92
    +0.10
    0.03
    Negative Logits
     unnecessarily
    -0.76
     prohib
    -0.75
     anecd
    -0.74
     potential
    -0.73
     robust
    -0.73
     liability
    -0.72
     generalized
    -0.72
     selective
    -0.71
     popul
    -0.71
     wholesale
    -0.71
    POSITIVE LOGITS
     Yourself
    1.12
     Tonight
    1.06
     Guys
    1.05
     Dating
    1.04
     Forever
    1.04
     Wrong
    1.03
     Them
    1.03
     Hate
    1.02
     Feeling
    1.02
     Love
    0.99
    Activations Density 0.086%

    No Known Activations