© 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
    3. Llama3.3-70B-IT
    5. 50-RESID-POST-GF
    7. 4943
    Prev
    Next
    INDEX
    Explanations

    performance degradation or unaffected

    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
     shrinking
    -0.12
     Hank
    -0.11
     dwind
    -0.10
     dying
    -0.10
    haps
    -0.09
     discour
    -0.09
     Grimm
    -0.09
     destructive
    -0.09
     discourage
    -0.09
    igs
    -0.08
    POSITIVE LOGITS
     degradation
    0.24
     ÑĥÑħ
    0.23
     sacrifice
    0.21
    Deg
    0.21
     performance
    0.20
     Deg
    0.20
     compromise
    0.19
     loss
    0.19
     sacrificed
    0.18
     degraded
    0.17
    Activations Density 0.083%

    No Known Activations