© 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. 1-RES-JB
    11. 22372
    Prev
    Next
    INDEX
    Explanations

    words related to taking defensive or opposing actions

    oai_token-act-pair · gpt-3.5-turbo

    references to counter actions or measures against a specific issue or threat

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    doms
    -0.80
     Forge
    -0.79
     ABE
    -0.70
     Known
    -0.70
     Become
    -0.68
     Extended
    -0.66
     Leaving
    -0.65
    OOD
    -0.64
     Mustang
    -0.62
     Coh
    -0.61
    POSITIVE LOGITS
    counter
    1.33
     counters
    1.07
    Counter
    1.06
     counter
    1.05
    balance
    1.01
    intuitive
    1.00
    tops
    0.97
    intelligence
    0.96
    balanced
    0.96
    attack
    0.95
    Activations Density 0.011%

    No Known Activations