© 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. 3980
    Prev
    Next
    INDEX
    Explanations

    words related to physical constraints or struggles, such as binding, isolation, or pushing into straits

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

    the term "stra," indicating a focus on variants of the word related to "straight" or "straightness."

    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
     Lyft
    -0.65
     ballots
    -0.64
     Ore
    -0.63
     delinquent
    -0.60
     rou
    -0.60
     seasons
    -0.60
     archived
    -0.59
     elig
    -0.58
     earth
    -0.57
     informational
    -0.57
    POSITIVE LOGITS
    pped
    1.35
    pping
    1.25
    fing
    1.22
    ights
    1.18
    ÃŁ
    1.14
    fe
    1.13
    ighter
    1.09
    ussian
    1.03
    ps
    1.03
    its
    1.01
    Activations Density 0.023%

    No Known Activations