© 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 24576
    9. 8-RES_FS24576-JB
    11. 12299
    Prev
    Next
    INDEX
    Explanations

    personal pronouns indicating actions or emotions directed towards oneself or others

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    139
    +0.12
    0.5%
    300
    +0.11
    0.5%
    675
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.12
    0.06
    477
    +0.11
    0.04
    300
    +0.10
    0.05
    Negative Logits
     Associated
    -0.68
    quartered
    -0.66
     Rousse
    -0.66
     Electrical
    -0.63
     Globe
    -0.62
    laughter
    -0.61
     Canaver
    -0.61
     understatement
    -0.61
     Stra
    -0.61
    holding
    -0.60
    POSITIVE LOGITS
    've
    1.10
     reached
    0.96
     arrived
    0.94
     got
    0.92
     realise
    0.92
     finally
    0.92
     realize
    0.91
    're
    0.90
     realized
    0.90
     reaches
    0.90
    Activations Density 0.145%

    No Known Activations