© 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. 9831
    Prev
    Next
    INDEX
    Explanations

    the word "Speaker" followed by a name or position

    oai_token-act-pair · gpt-3.5-turbo

    mentions of political figures, specifically speakers of the House

    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
    arial
    -0.80
    urat
    -0.72
     Belarus
    -0.68
     Mines
    -0.64
     Devils
    -0.62
     mines
    -0.58
    kus
    -0.58
    ties
    -0.57
    sters
    -0.57
     Mining
    -0.57
    POSITIVE LOGITS
    phone
    0.94
     Newt
    0.84
    Ryan
    0.83
     speaker
    0.83
     Boehner
    0.80
    NetMessage
    0.79
    woman
    0.78
     Gingrich
    0.77
    hip
    0.75
    person
    0.73
    Activations Density 0.050%

    No Known Activations