© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 5809
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or outcomes that can be considered negative or questionable

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases indicating negative consequences or cautionary terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     assemblies
    -0.88
     Emails
    -0.81
    assies
    -0.81
     interviews
    -0.81
     rows
    -0.81
     backgrounds
    -0.79
     interfaces
    -0.79
     galleries
    -0.78
     Tours
    -0.77
     reviews
    -0.77
    POSITIVE LOGITS
    brainer
    1.20
    starter
    1.18
    acea
    1.03
    breaker
    0.99
    shoot
    0.94
    requisite
    0.94
    lude
    0.94
    statement
    0.90
     breaker
    0.90
    product
    0.88
    Activations Density 0.230%

    No Known Activations