© 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. 26762
    Prev
    Next
    INDEX
    Explanations

    phrases indicating purpose or intent

    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
    718
    +0.13
    0.6%
    608
    +0.13
    0.6%
    665
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.13
    0.04
    608
    +0.13
    0.04
    665
    +0.10
    0.04
    Negative Logits
    heard
    -0.69
    CTV
    -0.68
    pointers
    -0.67
     Intermediate
    -0.66
     Creed
    -0.65
    estern
    -0.65
    backs
    -0.64
    listed
    -0.63
     Jem
    -0.63
    apps
    -0.62
    POSITIVE LOGITS
     undermine
    1.25
     amplify
    1.20
     diminish
    1.19
     alleviate
    1.18
     enhance
    1.17
     accelerate
    1.16
     lessen
    1.15
     strengthen
    1.12
     elevate
    1.11
     degrade
    1.11
    Activations Density 0.077%

    No Known Activations