© 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. 19118
    Prev
    Next
    INDEX
    Explanations

    phrases related to resistance or denial

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "refusing" and its variations, indicating a focus on acts of non-compliance or rejection

    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
    ussion
    -0.67
    Cam
    -0.66
     impacts
    -0.66
     aug
    -0.64
    Sum
    -0.63
    impact
    -0.62
     Impact
    -0.62
     upl
    -0.61
    ather
    -0.61
    inav
    -0.60
    POSITIVE LOGITS
     refusing
    3.12
     refusal
    2.38
     refuse
    2.12
     refuses
    2.08
     refused
    2.03
     insisting
    1.77
     rejecting
    1.77
     denying
    1.71
     unwilling
    1.61
     resisting
    1.56
    Activations Density 0.025%

    No Known Activations