Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 17170
    Prev
    Next
    INDEX
    Explanations

    the word "Right."

    oai_token-act-pair · gpt-3.5-turbo

    the term "Right" as it pertains to political ideology

    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
    TABLE
    -0.58
    ux
    -0.56
    eland
    -0.56
    agos
    -0.54
    iatrics
    -0.54
    lled
    -0.54
    gypt
    -0.53
    RESULTS
    -0.52
     culmin
    -0.52
    ulence
    -0.51
    POSITIVE LOGITS
     Right
    3.63
    Right
    2.91
     RIGHT
    2.40
     right
    2.15
    right
    2.12
     Left
    1.72
     Wrong
    1.66
    Left
    1.57
     Correct
    1.39
     Rights
    1.31
    Activations Density 0.013%

    No Known Activations