© 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. 11687
    Prev
    Next
    INDEX
    Explanations

    words related to analysis and evaluation, especially terms indicating assessment or judgment such as "meaningful," "practical," and "credible."

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing skepticism or questioning the validity of claims

    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
    cellent
    -0.79
    mostly
    -0.77
     nonetheless
    -0.72
     Initially
    -0.72
    olutely
    -0.71
     Mostly
    -0.71
    almost
    -0.68
     briefly
    -0.68
    sometimes
    -0.67
    Simple
    -0.65
    POSITIVE LOGITS
     whatsoever
    1.30
     nor
    1.20
     explanation
    1.13
     repercussions
    1.04
     reperc
    1.02
     indication
    1.00
     reason
    1.00
     recourse
    0.99
     connection
    0.97
     affiliation
    0.96
    Activations Density 0.202%

    No Known Activations