© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 8482
    Prev
    Next
    INDEX
    Explanations

    descriptions of criminal activities

    oai_token-act-pair · gpt-3.5-turbo

    signs of conflict or demand in dramatic situations, particularly involving firearms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

    Negative Logits
     Printed
    -0.74
     Brilliant
    -0.72
    isite
    -0.69
     inval
    -0.69
     superb
    -0.68
     dystop
    -0.67
     seeded
    -0.66
     Quarterly
    -0.66
    etheless
    -0.66
     archives
    -0.65
    POSITIVE LOGITS
     incrim
    0.83
     explosives
    0.81
     tresp
    0.80
     safety
    0.78
     robbery
    0.78
     intruder
    0.77
     evacuation
    0.77
     burglary
    0.76
     police
    0.76
     evacuate
    0.76
    Activations Density 0.626%

    No Known Activations