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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10192
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to superlatives, such as "most" and "highly"

    oai_token-act-pair · gpt-3.5-turbo

    the term "most" in various contexts, indicating comparison or superlative qualities

    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
    idas
    -0.80
     Desk
    -0.79
     Extensions
    -0.75
    reads
    -0.74
    uggest
    -0.73
     Anniversary
    -0.72
    estyles
    -0.71
    odies
    -0.71
    phies
    -0.70
    oki
    -0.70
    POSITIVE LOGITS
     populous
    1.50
     populated
    1.41
     industrialized
    1.34
     densely
    1.33
     affluent
    1.22
     crowded
    1.22
     prosperous
    1.21
     mountainous
    1.21
     congest
    1.18
     polluted
    1.14
    Activations Density 0.190%

    No Known Activations