© 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. 9456
    Prev
    Next
    INDEX
    Explanations

    adjectives that indicate correctness or appropriateness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "proper" in various contexts

    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
    oub
    -0.68
    hunt
    -0.65
    Grab
    -0.63
     Deals
    -0.63
    apeshifter
    -0.62
    whel
    -0.60
    chant
    -0.60
    onite
    -0.59
    esters
    -0.59
     Despair
    -0.58
    POSITIVE LOGITS
     proper
    3.71
     Proper
    2.41
     properly
    1.96
     appropriate
    1.88
     correct
    1.73
     adequate
    1.56
     rightful
    1.54
     requisite
    1.36
     suitable
    1.33
     improper
    1.33
    Activations Density 0.011%

    No Known Activations