© 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. 4-RES-JB
    11. 19603
    Prev
    Next
    INDEX
    Explanations

    mentions of paying attention or lack thereof

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the act of paying attention

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

    No Comments

    Negative Logits
     halves
    -0.74
    tre
    -0.71
    lua
    -0.65
     Zamb
    -0.64
     Zi
    -0.64
     Mehran
    -0.63
    ft
    -0.62
    iche
    -0.62
    thing
    -0.60
     hunt
    -0.59
    POSITIVE LOGITS
    arios
    0.85
    ibly
    0.82
    othal
    0.79
     attent
    0.75
    endment
    0.73
    ibility
    0.73
    IBLE
    0.72
     attentive
    0.72
    é¾įå¥ij士
    0.71
     attention
    0.71
    Activations Density 0.012%

    No Known Activations