© 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. 10-RES-JB
    11. 16932
    Prev
    Next
    INDEX
    Explanations

    comparative relationships such as "more than" and "better than."

    oai_token-act-pair · gpt-3.5-turbo

    conditional phrases suggesting choices or alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Illustrated
    -0.80
    omen
    -0.80
    iverse
    -0.79
    erning
    -0.73
    htaking
    -0.72
    ãĤ´
    -0.72
    eatures
    -0.72
     Beaut
    -0.72
     Medals
    -0.71
    abase
    -0.69
    POSITIVE LOGITS
     surrender
    1.50
     obey
    1.40
     withdraw
    1.39
     acquies
    1.37
     retaliate
    1.37
     cooperate
    1.36
     refuse
    1.35
     flee
    1.35
     resign
    1.31
     relent
    1.30
    Activations Density 0.325%

    No Known Activations