© 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. 10-RES-JB
    11. 6492
    Prev
    Next
    INDEX
    Explanations

    phrases expressing intention or commitment

    oai_token-act-pair · gpt-3.5-turbo

    statements related to commitment, accountability, and collective action

    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
    atown
    -0.67
     Ago
    -0.63
    aukee
    -0.61
    throp
    -0.61
    ikk
    -0.60
    Abstract
    -0.59
    cture
    -0.59
     Skydragon
    -0.58
     Fold
    -0.58
    cohol
    -0.57
    POSITIVE LOGITS
     accordingly
    1.52
     consequently
    1.13
     therefore
    0.95
     rightfully
    0.94
     certainly
    0.92
     consequ
    0.91
     hopefully
    0.88
     gladly
    0.87
     therein
    0.87
     thereof
    0.86
    Activations Density 0.478%

    No Known Activations