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

    Neuronpedia

    Jacobian LensNEW
    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. 43056
    Prev
    Next
    INDEX
    Explanations

    instances of the word "accept" in various contexts

    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
    183
    +0.13
    0.6%
    339
    +0.12
    0.6%
    654
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    654
    +0.13
    0.02
    183
    +0.12
    0.02
    236
    +0.12
    0.02
    Negative Logits
    ionage
    -0.77
    borgh
    -0.76
    brow
    -0.74
    vertisement
    -0.70
    ixt
    -0.69
    fox
    -0.69
    iferation
    -0.67
     glands
    -0.67
     Indust
    -0.67
    borough
    -0.66
    POSITIVE LOGITS
     responsibility
    1.10
     defeat
    0.94
    ably
    0.90
     accept
    0.84
     acceptance
    0.83
     uncond
    0.81
     accepting
    0.81
     bribes
    0.80
    enance
    0.78
     rejection
    0.78
    Activations Density 0.016%

    No Known Activations