© 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. 2-RES-JB
    11. 8942
    Prev
    Next
    INDEX
    Explanations

    terms related to harmful or dangerous individuals or entities

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "killer" in various contexts, which likely indicates a focus on terms associated with dangerous entities or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rity
    -0.89
    bles
    -0.80
    urn
    -0.77
    ational
    -0.77
    ional
    -0.77
    edu
    -0.74
    ured
    -0.72
    ratulations
    -0.71
    rir
    -0.70
    urat
    -0.69
    POSITIVE LOGITS
     killer
    1.14
    killer
    1.05
     killers
    0.97
     Killer
    0.91
     spree
    0.84
     whales
    0.82
     whale
    0.76
    knife
    0.75
     beware
    0.72
    fish
    0.71
    Activations Density 0.013%

    No Known Activations