© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 10687
    Prev
    Next
    INDEX
    Explanations

    verbs indicating action or decision

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the concept of "taking" or "not taking" something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lehem
    -0.70
    naire
    -0.69
    alam
    -0.67
    éŃĶ
    -0.67
    icol
    -0.66
    fw
    -0.66
    女
    -0.65
    lex
    -0.65
    ingen
    -0.64
    nell
    -0.64
    POSITIVE LOGITS
     kindly
    1.16
     advantage
    0.96
     anymore
    0.94
     lightly
    0.88
     responsibility
    0.86
     seriously
    0.86
     sides
    0.85
     aback
    0.82
    aways
    0.81
     heed
    0.79
    Activations Density 0.049%

    No Known Activations