© 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. 9-RES-JB
    11. 3604
    Prev
    Next
    INDEX
    Explanations

    words related to appealing or targeting specific audiences

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to audience appeal or target demographics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aping
    -0.72
     laun
    -0.72
    work
    -0.68
    together
    -0.66
    facing
    -0.66
    cknowled
    -0.65
    conduct
    -0.65
    speak
    -0.64
    run
    -0.64
    practice
    -0.63
    POSITIVE LOGITS
     sensibilities
    1.21
     curiosity
    1.06
     audiences
    0.93
     appetite
    0.89
     voters
    0.83
     craving
    0.83
     palate
    0.83
     itch
    0.83
     viewers
    0.82
    ravings
    0.82
    Activations Density 0.209%

    No Known Activations