© 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. 7-RES-JB
    11. 3154
    Prev
    Next
    INDEX
    Explanations

    violent actions or incidents involving weapons or explosives

    oai_token-act-pair · gpt-3.5-turbo

    mentions of violent events and attacks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĤ´ãĥ³
    -0.92
    xus
    -0.83
    natureconservancy
    -0.83
    bred
    -0.78
    proxy
    -0.74
    }}}
    -0.74
    URI
    -0.73
     fixme
    -0.72
    olkien
    -0.72
    cius
    -0.71
    POSITIVE LOGITS
     unarmed
    1.16
     targets
    0.95
    bomb
    0.93
     indiscrim
    0.90
     deton
    0.87
     wounding
    0.86
     bombing
    0.84
     civilians
    0.83
     spree
    0.82
     civilian
    0.80
    Activations Density 0.235%

    No Known Activations