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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Llama3.3-70B-IT
    5. 50-RESID-POST-GF
    7. 630
    Prev
    Next
    INDEX
    Explanations

    best, ideal, keywords

    np_acts-logits-general · gemini-2.5-flash-lite

    phrases that frame or discuss selecting the “best” or ideal option, often in question-answer or explanatory setup language.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.3-70B-IT @ 50-resid-post-gf
    Configuration
    Goodfire/Llama-3.3-70B-Instruct-SAE-l50/Llama-3.3-70B-Instruct-SAE-l50.pt
    Prompts (Dashboard)
    10,000 prompts, 128 tokens each
    Dataset (Dashboard)
    lmsys/lmsys-chat-1m
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     widest
    -0.10
     weakest
    -0.10
     Newest
    -0.10
     youngest
    -0.09
     slightest
    -0.09
     simplest
    -0.09
     darkest
    -0.09
     POSS
    -0.08
     oldest
    -0.08
    oria
    -0.08
    POSITIVE LOGITS
     best
    0.58
    æľĢä½³
    0.50
    best
    0.43
    -best
    0.40
     beste
    0.38
     лÑĥÑĩ
    0.37
     BEST
    0.37
    _best
    0.36
     ideal
    0.36
     Best
    0.36
    Activations Density 0.398%

    No Known Activations