© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8-RES-JB
    11. 4230
    Prev
    Next
    INDEX
    Explanations

    comparative terms indicating judgment or evaluation, such as "better" or "worse"

    oai_token-act-pair · gpt-3.5-turbo

    comparative adjectives that express improvement or decline

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    esville
    -0.74
    ettes
    -0.72
     Gazette
    -0.70
    MG
    -0.65
    gemony
    -0.65
    SEA
    -0.64
    NH
    -0.64
    essee
    -0.64
    Nob
    -0.63
    trl
    -0.63
    POSITIVE LOGITS
     than
    1.06
     Than
    1.01
    than
    0.90
     behaved
    0.88
     suited
    0.82
     eleph
    0.79
     acquainted
    0.68
     vers
    0.68
    iation
    0.68
     bang
    0.68
    Activations Density 0.049%

    No Known Activations