© 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. 0-RES-JB
    11. 4112
    Prev
    Next
    INDEX
    Explanations

    mentions of the politician Marco Rubio

    oai_token-act-pair · gpt-3.5-turbo

    mentions of political figures, specifically Marco Rubio

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     priv
    -0.73
     WW
    -0.73
     oath
    -0.71
     industrial
    -0.69
     Legion
    -0.68
    hern
    -0.65
     eth
    -0.65
     arrow
    -0.65
     undert
    -0.64
     emb
    -0.64
    POSITIVE LOGITS
     Rubio
    3.86
    Rub
    2.01
    Cruz
    1.83
     Kasich
    1.81
     Cruz
    1.79
     Marco
    1.74
    Marco
    1.61
     Fiorina
    1.60
     Santorum
    1.59
     Huckabee
    1.57
    Activations Density 0.017%

    No Known Activations