© 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. 1-RES-JB
    11. 5128
    Prev
    Next
    INDEX
    Explanations

    instances of rejection or disapproval

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "rejected."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RH
    -0.77
    cycl
    -0.77
    arist
    -0.77
    Bio
    -0.72
    ipel
    -0.71
    opter
    -0.71
    dr
    -0.70
    istics
    -0.69
    ecycle
    -0.67
    oppy
    -0.67
    POSITIVE LOGITS
     unanimously
    0.92
     rejected
    0.91
     reject
    0.80
     rejecting
    0.77
     outright
    0.76
    Ĥª
    0.76
     rejection
    0.72
     warr
    0.69
     rejects
    0.67
     validation
    0.66
    Activations Density 0.010%

    No Known Activations