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. 12-RES-JB
    6. 17814
    Prev
    Next
    INDEX
    Explanations

    words related to politics, specifically transitioning or moving forward in a political context

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    actions and verbs related to movement and political transitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
     altogether
    -0.59
    unless
    -0.58
    misc
    -0.57
     Wrong
    -0.57
    nox
    -0.56
     Anything
    -0.56
     kidding
    -0.56
    ãĤ±
    -0.55
    otally
    -0.55
    è£
    -0.55
    POSITIVE LOGITS
     closer
    0.91
     nearer
    0.88
     unfolds
    0.87
     toward
    0.87
     deeper
    0.87
     towards
    0.81
     prepares
    0.79
     tighter
    0.73
     unfold
    0.72
     onward
    0.71
    Activations Density 0.277%

    No Known Activations