© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 10680
    Prev
    Next
    INDEX
    Explanations

    the word "wanted" followed by an infinitive verb

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating intent or desire to take action

    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
    Not in Any Lists

    No Comments

    Negative Logits
     presently
    -0.70
    lies
    -0.67
    aka
    -0.66
     Rapids
    -0.65
     currently
    -0.64
     ongoing
    -0.63
    requires
    -0.63
    soever
    -0.61
    ested
    -0.61
    currently
    -0.60
    POSITIVE LOGITS
     capitalize
    1.18
     explore
    1.04
     recreate
    1.04
     maximize
    1.04
     eliminate
    1.03
     postpone
    1.02
     introduce
    1.01
     emphasize
    1.01
     create
    0.99
     formulate
    0.99
    Activations Density 0.246%

    No Known Activations