© 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. 9-RES-JB
    11. 15805
    Prev
    Next
    INDEX
    Explanations

    terms related to cross-promotion

    oai_token-act-pair · gpt-3.5-turbo

    references to cross-disciplinary or cross-functional themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Lump
    -0.67
     sorely
    -0.64
     Vac
    -0.63
     Mons
    -0.62
     Pitch
    -0.62
     CGI
    -0.59
     Pretty
    -0.58
     Brave
    -0.58
     Mad
    -0.58
     payday
    -0.58
    POSITIVE LOGITS
    sectional
    1.28
    border
    1.14
    examination
    1.13
    disciplinary
    1.10
    country
    1.10
    section
    1.08
    cultural
    1.05
    platform
    1.03
    continental
    1.03
    functional
    1.02
    Activations Density 0.024%

    No Known Activations