© 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. 10-RES-JB
    11. 2960
    Prev
    Next
    INDEX
    Explanations

    superlatives, such as "most," indicating high levels or extremes within a comparison set

    oai_token-act-pair · gpt-3.5-turbo

    phrases that rank or describe something as "one of the most" in a particular context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    instead
    -0.87
    runs
    -0.80
    æ©
    -0.80
     Loans
    -0.80
    ulia
    -0.76
    pta
    -0.75
    eto
    -0.74
    anges
    -0.74
    heid
    -0.74
    EStream
    -0.73
    POSITIVE LOGITS
     recognizable
    1.27
     exciting
    1.25
     important
    1.21
     underrated
    1.18
     intriguing
    1.18
     prolific
    1.17
     influential
    1.16
     misunderstood
    1.14
     interesting
    1.14
     fascinating
    1.13
    Activations Density 0.057%

    No Known Activations