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

    Neuronpedia

    Jacobian LensNEW
    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. 12716
    Prev
    Next
    INDEX
    Explanations

    phrases related to advocating or endorsing something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "promoting."

    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
    iaries
    -0.70
    ould
    -0.69
    ISE
    -0.69
    ohl
    -0.60
    raid
    -0.60
    ise
    -0.59
     follow
    -0.59
    Mine
    -0.59
    enne
    -0.59
    url
    -0.59
    POSITIVE LOGITS
     promoting
    2.59
     advocating
    2.24
     criticizing
    2.10
     denouncing
    2.05
     endorsing
    2.01
     portraying
    1.89
     upholding
    1.88
     condemning
    1.84
     protesting
    1.83
     praising
    1.82
    Activations Density 0.056%

    No Known Activations