© 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. 11-RES-JB
    11. 12605
    Prev
    Next
    INDEX
    Explanations

    phrases related to the evaluation of importance or judgment

    oai_token-act-pair · gpt-3.5-turbo

    instances of the article "a" followed by various contexts or topics

    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
    Types
    -0.76
    igans
    -0.74
    rates
    -0.72
    loads
    -0.71
    uploads
    -0.67
    chers
    -0.67
    effects
    -0.65
    almost
    -0.65
    Contents
    -0.64
    rils
    -0.63
    POSITIVE LOGITS
     particular
    1.27
     given
    1.10
     certain
    1.07
     specific
    0.98
     competitor
    0.92
     colleague
    0.87
     person
    0.87
     desired
    0.86
     stranger
    0.86
     piece
    0.85
    Activations Density 0.317%

    No Known Activations