© 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. 10-RES-JB
    11. 7800
    Prev
    Next
    INDEX
    Explanations

    phrases related to planned operations or events carried out under the cover of darkness

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to social and political issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     creep
    -0.84
     ignor
    -0.78
     lungs
    -0.76
     excel
    -0.76
     guts
    -0.75
     screw
    -0.74
     reserves
    -0.74
     optimized
    -0.74
     claws
    -0.73
     enriched
    -0.73
    POSITIVE LOGITS
    Meanwhile
    1.59
    According
    1.55
    Speaking
    1.54
    Despite
    1.53
    Earlier
    1.51
    During
    1.50
    While
    1.49
    Although
    1.46
    Another
    1.46
    Among
    1.44
    Activations Density 0.084%

    No Known Activations