© 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. 22124
    Prev
    Next
    INDEX
    Explanations

    instances where something is deemed insufficient or lacking

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "enough" and similar expressions regarding sufficiency

    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
    Ħ¢
    -0.68
    gha
    -0.68
    Ń·
    -0.67
    midt
    -0.61
    aming
    -0.61
    gdala
    -0.61
    yrinth
    -0.60
    asus
    -0.60
    iries
    -0.60
    ggles
    -0.59
    POSITIVE LOGITS
     motivation
    0.97
     reason
    0.94
     justification
    0.94
     impetus
    0.93
     incentive
    0.90
     proof
    0.90
     firepower
    0.88
     evidence
    0.84
     to
    0.83
     deterrent
    0.82
    Activations Density 0.047%

    No Known Activations