© 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. 11-RES-JB
    11. 1302
    Prev
    Next
    INDEX
    Explanations

    language related to negative emotions and conflict

    oai_token-act-pair · gpt-3.5-turbo

    themes of fear, intimidation, and societal issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uve
    -0.77
     Pets
    -0.77
    entials
    -0.74
    arate
    -0.74
    pps
    -0.74
    audi
    -0.73
    perties
    -0.72
    Newsletter
    -0.72
    adelphia
    -0.71
     Railway
    -0.71
    POSITIVE LOGITS
     uncertainty
    1.46
     confusion
    1.43
     instability
    1.38
     insecurity
    1.37
     frustration
    1.37
     upheaval
    1.35
     dysfunction
    1.34
     bloodshed
    1.33
     hardship
    1.32
     misinformation
    1.31
    Activations Density 0.188%

    No Known Activations