© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35258
    Prev
    Next
    INDEX
    Explanations

    almost + quantifier/adverb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_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
    -0.09
    å°½éĩı
    -0.09
    0
    -0.09
    èĩ³å°ij
    -0.09
    è¦ģä¸įçĦ¶
    -0.09
    åıįæŃ£
    -0.08
    8
    -0.08
    resse
    -0.08
    stakes
    -0.08
    éķ¿çĽ¸
    -0.08
    POSITIVE LOGITS
    æĬ¡
    0.10
    åģļåĪ°
    0.09
    åħ¨èº«
    0.09
    åħ¨é¢Ŀ
    0.09
    åħ¨æĺ¯
    0.09
    è¿Ľå±ķ
    0.09
    夺
    0.09
    æ¯ı次éĥ½
    0.09
    褫
    0.09
    å²Ń
    0.08
    Activations Density 0.037%

    No Known Activations