© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9847
    Prev
    Next
    INDEX
    Explanations

    words related to extreme behavior or actions, as well as aspects related to politeness and manners

    oai_token-act-pair · gpt-3.5-turbo

    extreme behaviors and their impact

    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
     Started
    -0.75
     Timeline
    -0.67
     Completed
    -0.67
     Previous
    -0.65
     CM
    -0.65
     Prior
    -0.63
     Updated
    -0.62
     Announce
    -0.60
     Nationwide
    -0.58
     Streaming
    -0.58
    POSITIVE LOGITS
     slightest
    0.79
     lest
    0.76
    sometimes
    0.70
     occasionally
    0.69
     occasional
    0.67
     even
    0.67
    anything
    0.66
    azor
    0.64
    ifling
    0.64
    udder
    0.63
    Activations Density 0.675%

    No Known Activations