© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44163
    Prev
    Next
    INDEX
    Explanations

    Overcorrection

    np_max-act · gemini-2.0-flash

    The neuron fires on words that signal overcorrection or going to extremes—terms like “tipping too far,” “overcorrect,” “overcompensating,” etc.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    asion
    -0.07
    .sg
    -0.07
    "If
    -0.06
    rupa
    -0.06
    Noise
    -0.06
     rootView
    -0.06
    out
    -0.06
    .bed
    -0.06
    ordon
    -0.06
     oznám
    -0.06
    POSITIVE LOGITS
     Fleet
    0.07
    آخر
    0.06
    	RTCT
    0.06
     ني
    0.06
    $new
    0.06
    _learning
    0.06
     чем
    0.06
    zyst
    0.06
    huge
    0.06
     khám
    0.06
    Activations Density 0.030%

    No Known Activations