© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1-RES-JB
    11. 22721
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "liberals" in a context that includes either positive or negative connotations

    oai_token-act-pair · gpt-3.5-turbo

    references to liberals and progressives in a political context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lying
    -0.75
     Sul
    -0.74
    teen
    -0.70
    lly
    -0.68
     mineral
    -0.65
    aring
    -0.65
     volcanic
    -0.64
     maternal
    -0.63
    Gi
    -0.63
    yss
    -0.61
    POSITIVE LOGITS
    aurus
    1.11
    ervatives
    1.06
     Liberals
    0.91
    ervative
    0.87
     liberals
    0.84
    arians
    0.84
    umenthal
    0.82
    ists
    0.81
    rats
    0.79
    paces
    0.78
    Activations Density 0.009%

    No Known Activations