© 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 EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 3994
    Prev
    Next
    INDEX
    Explanations

    phrases related to unwavering commitment or resistance

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    negative prefixes indicating an absence or lack of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
     mileage
    -0.69
     steroids
    -0.68
     hemor
    -0.68
     hygiene
    -0.65
     Bots
    -0.64
     rumors
    -0.60
     blot
    -0.59
     moot
    -0.59
     altern
    -0.59
     Role
    -0.58
    POSITIVE LOGITS
    avering
    1.84
    arranted
    1.76
    ritten
    1.67
    ield
    1.60
    inding
    1.52
    inn
    1.36
    isdom
    1.32
    atched
    1.22
    itness
    1.21
    atch
    1.20
    Activations Density 0.011%

    No Known Activations