© 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. 20361
    Prev
    Next
    INDEX
    Explanations

    phrases related to business and success

    oai_token-act-pair · gpt-3.5-turbo

    phrases that discuss success in a professional context

    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
    ury
    -0.76
    oute
    -0.72
    ãĤ§
    -0.67
    Minor
    -0.64
     lamented
    -0.64
     remainder
    -0.62
     wonders
    -0.62
    asar
    -0.62
    holiday
    -0.61
    oday
    -0.61
    POSITIVE LOGITS
     sufficiently
    1.01
     yourself
    1.00
     oneself
    0.89
     willing
    0.85
     believable
    0.82
     intimately
    0.81
     KNOW
    0.81
     articulate
    0.80
     beforehand
    0.79
     premise
    0.79
    Activations Density 0.652%

    No Known Activations