© 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. 0-RES-JB
    11. 19187
    Prev
    Next
    INDEX
    Explanations

    the word "agree" followed by a number that indicates the strength of agreement

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to consensus or agreement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Spotlight
    -0.60
    lot
    -0.60
    bound
    -0.60
     Decay
    -0.59
    flying
    -0.59
     Shattered
    -0.59
    oline
    -0.57
     Hast
    -0.57
     Crush
    -0.56
     Haunted
    -0.56
    POSITIVE LOGITS
     agree
    3.61
     agrees
    2.42
    agree
    2.15
     disagree
    2.05
     agreeing
    2.01
     agreed
    1.99
     approve
    1.62
     concede
    1.60
     agre
    1.50
     admit
    1.49
    Activations Density 0.012%

    No Known Activations