© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12419
    Prev
    Next
    INDEX
    Explanations

    he followed by past tense verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    å®ĥ
    -0.27
     need
    -0.24
    éľĢè¦ģçĶ¨
    -0.22
    å®ĥçļĦ
    -0.22
    éľĢè¦ģ
    -0.22
    åĪ°æŶåĢĻ
    -0.21
    [Unit
    -0.21
    让å®ĥ
    -0.21
     themselves
    -0.20
     gotta
    -0.20
    POSITIVE LOGITS
    å¤ļ次
    0.24
    æĽ¾
    0.21
    éĩį度
    0.21
     пеÑĢе
    0.20
    ä¹ŁæĽ¾
    0.20
    被èªī为
    0.20
    两次
    0.20
    æĽ¾ä»»
    0.19
    endors
    0.19
    undef
    0.19
    Activations Density 0.147%

    No Known Activations