© 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 24576
    9. 8-RES_FS24576-JB
    11. 18503
    Prev
    Next
    INDEX
    Explanations

    punctuation marks at the end of sentences

    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
    373
    +0.12
    0.6%
    626
    +0.11
    0.5%
    37
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    626
    +0.12
    0.02
    304
    +0.11
    0.02
    594
    +0.09
    0.01
    Negative Logits
     deceptive
    -0.61
     confidentiality
    -0.57
     whistleblowers
    -0.57
     privacy
    -0.57
     outgoing
    -0.56
     safety
    -0.56
     advis
    -0.55
     confidential
    -0.54
     trusted
    -0.54
     stewards
    -0.53
    POSITIVE LOGITS
    imgur
    1.01
    e
    0.91
    aca
    0.75
    hat
    0.75
    seed
    0.74
    MX
    0.71
    hs
    0.71
    minimum
    0.71
    ¼
    0.70
    medium
    0.70
    Activations Density 0.021%

    No Known Activations