© 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. 2720
    Prev
    Next
    INDEX
    Explanations

    passive constructions with "are" as the main verb

    oai_token-act-pair · gpt-3.5-turbo

    phrases or structures that indicate the presence of the word "are" significantly used in various contexts

    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
    iliation
    -0.76
    ename
    -0.74
    isition
    -0.73
    usp
    -0.71
    matter
    -0.70
    illation
    -0.68
    fleet
    -0.68
    lee
    -0.67
    escription
    -0.67
    comb
    -0.67
    POSITIVE LOGITS
     plenty
    1.22
     countless
    1.11
     exceptions
    1.10
     lots
    1.06
     many
    1.05
     innumerable
    1.05
     occasions
    1.05
     myriad
    1.02
     dozens
    1.00
     certain
    0.99
    Activations Density 0.085%

    No Known Activations