© 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. 19024
    Prev
    Next
    INDEX
    Explanations

    statements related to public opinion and health assessments

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating public opinion ratings or assessments

    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
     jewels
    -0.76
    perial
    -0.72
     Normandy
    -0.69
     dyn
    -0.68
     carving
    -0.67
     ILCS
    -0.66
     jewel
    -0.66
     Rampage
    -0.65
     Grail
    -0.64
    gart
    -0.64
    POSITIVE LOGITS
     Unsure
    1.45
     undecided
    1.44
    ecided
    1.32
    agree
    1.24
    yes
    1.24
     unsure
    1.22
     yes
    1.22
     Somew
    1.20
    Yes
    1.19
     YES
    1.19
    Activations Density 0.290%

    No Known Activations