© 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. 24074
    Prev
    Next
    INDEX
    Explanations

    mentions of disclosure requirements about personal and public interests related to government positions

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to legal and ethical obligations in a governance context

    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
    ãĥ¥
    -0.77
     Clockwork
    -0.72
    essim
    -0.71
    ursday
    -0.68
    crim
    -0.68
     Survive
    -0.67
    terness
    -0.67
    »Ĵ
    -0.66
     Symptoms
    -0.65
    ptoms
    -0.65
    POSITIVE LOGITS
     dealings
    1.21
     influence
    1.20
     lobbying
    1.19
     lobbyists
    1.12
     lobbyist
    1.11
     consulting
    1.10
     appoint
    1.09
     consultancy
    1.09
    orship
    1.09
     expenditures
    1.07
    Activations Density 0.492%

    No Known Activations