© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 13388
    Prev
    Next
    INDEX
    Explanations

    an followed by specific nouns

    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
    ingly
    -0.22
    ðŁı»
    -0.20
    ëĬĶ
    -0.19
    empre
    -0.18
    illance
    -0.18
    иÑħ
    -0.17
    igner
    -0.16
    iest
    -0.16
    leur
    -0.16
    eren
    -0.16
    POSITIVE LOGITS
    r
    0.21
    ivil
    0.20
    oxic
    0.19
    ãģĹãģ¾ãģĨ
    0.19
    agram
    0.19
    iversary
    0.19
    vil
    0.18
    tr
    0.18
    lies
    0.18
    vironment
    0.18
    Activations Density 0.356%

    No Known Activations