© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Llama3.3-70B-IT
    3. 50-RESID-POST-GF
    4. 58147
    Prev
    Next
    INDEX
    Explanations

    explanation of how"}, "how does back propergation"}, "called "Masked LM Pre"}, "fine-tune a"}, "a pre-trained language"}, "zero-shot also shifts"}The most prominent pattern seems to be describing how something is done or what a technique is called. how to explain

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     regress
    -0.10
    Regression
    -0.10
    çªģ
    -0.09
     poc
    -0.09
     Regression
    -0.09
    794
    -0.09
    '´
    -0.09
    æºĸ
    -0.09
     BoxFit
    -0.08
    .storage
    -0.08
    POSITIVE LOGITS
     ensemble
    0.14
     unders
    0.13
     Ensemble
    0.13
     Drop
    0.12
     techniques
    0.12
    Drop
    0.12
     Freeze
    0.12
     uns
    0.12
     dropout
    0.12
     curriculum
    0.11
    Activations Density 0.040%

    No Known Activations