Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 4853
    Prev
    Next
    INDEX
    Explanations

    phrases related to criminal activities and police investigations

    oai_token-act-pair · gpt-3.5-turbo

    legal or criminal-related terms and events

    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
    •single token firings
    Individually fires strongly on " indicted", " arrests", " Authorities", and other criminal-related terms.

    No Comments

    Negative Logits
     prest
    -0.66
     Niet
    -0.66
     nostalg
    -0.65
    Interstitial
    -0.62
    nat
    -0.62
    Reason
    -0.59
     vulner
    -0.59
     hereafter
    -0.59
     Moroc
    -0.58
    eem
    -0.58
    POSITIVE LOGITS
     delays
    0.72
    roversial
    0.66
     alleges
    0.66
    CLUS
    0.66
     wildfires
    0.66
    CLOSE
    0.65
     gunman
    0.65
     Investig
    0.65
     airstrikes
    0.65
     airstrike
    0.65
    Activations Density 0.685%

    No Known Activations