© 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. 10-RES-JB
    11. 18752
    Prev
    Next
    INDEX
    Explanations

    comparisons of different states or conditions

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases indicating change over time

    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
     marqu
    -0.59
    )].
    -0.56
    rf
    -0.56
     roll
    -0.52
    skirts
    -0.52
    gar
    -0.51
    erous
    -0.51
    Phill
    -0.51
    rich
    -0.51
    ouse
    -0.51
    POSITIVE LOGITS
     before
    1.01
     prior
    0.95
     days
    0.93
    before
    0.93
     years
    0.90
     ages
    0.90
    years
    0.85
     YEAR
    0.82
     BEFORE
    0.79
     previously
    0.79
    Activations Density 0.163%

    No Known Activations