© 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. 11-RES-JB
    11. 20616
    Prev
    Next
    INDEX
    Explanations

    verbs or noun phrases related to goals or intentions

    oai_token-act-pair · gpt-3.5-turbo

    statements about project goals or purposes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    soever
    -0.65
    soType
    -0.56
    avin
    -0.56
    vre
    -0.56
    pered
    -0.56
    regular
    -0.56
    ston
    -0.56
    vae
    -0.55
    bia
    -0.54
    oug
    -0.54
    POSITIVE LOGITS
     to
    1.14
     simple
    0.85
    to
    0.83
     maximizing
    0.77
     simply
    0.77
     preservation
    0.76
     simplicity
    0.75
     To
    0.74
     ensuring
    0.73
     TO
    0.72
    Activations Density 0.133%

    No Known Activations