© 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. 3633
    Prev
    Next
    INDEX
    Explanations

    phrases indicating intense focus or gaze on something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "staring" and its variations in the context of observing or focusing intently on something

    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
    accompan
    -0.81
    ocobo
    -0.72
    iHUD
    -0.69
    algia
    -0.64
     Recomm
    -0.60
    ItemTracker
    -0.59
    ãģ¦
    -0.58
    FER
    -0.58
     Provided
    -0.58
    notes
    -0.57
    POSITIVE LOGITS
     blank
    1.18
     dag
    1.12
     attent
    0.98
     curiously
    0.94
     downwards
    0.90
     anx
    0.89
     squarely
    0.88
     lov
    0.87
     downward
    0.86
     sideways
    0.86
    Activations Density 0.063%

    No Known Activations