© 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. Olmo-3-1125-32B
    3. 32-RES-BATCHTOPK-131K
    4. 46716
    Prev
    Next
    INDEX
    Explanations

    positive review adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    bcywinski/Olmo-3-32B-Base-SAE/saes_allenai_Olmo-3-1125-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     moot
    -0.11
     neighb
    -0.11
     Decorating
    -0.11
     indign
    -0.10
     Thing
    -0.10
    .which
    -0.10
     enraged
    -0.10
     blat
    -0.10
    )set
    -0.10
     stray
    -0.10
    POSITIVE LOGITS
     excellent
    0.22
     versatile
    0.20
     highly
    0.20
     Excellent
    0.19
    Excellent
    0.18
     Highly
    0.18
     extremely
    0.17
     powerful
    0.16
     excelente
    0.16
     strong
    0.16
    Activations Density 0.109%

    No Known Activations