Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 21849
    Prev
    Next
    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    chemical compounds associated with their upstream and downstream production processes.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    chemical compounds and their related synthesis processes.

    oai_token-act-pair · gpt-4oTriggered by @xinyanhu8

    The neuron responds to numeric tokens—especially the position-numbers in chemical names (e.g. “2,” “3,” “4,” “2,6,” etc.)—i.e. digits indicating substituent positions in compound nomenclature.

    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
     VERSION
    -0.07
     llev
    -0.06
    σί
    -0.06
    _HERSHEY
    -0.06
    /=
    -0.06
    	RTE
    -0.06
     hum
    -0.06
    enden
    -0.06
    �
    -0.06
     Vince
    -0.06
    POSITIVE LOGITS
     zeros
    0.07
    /shared
    0.07
    .theta
    0.07
    	pc
    0.06
     chemicals
    0.06
    hex
    0.06
     Bomb
    0.06
    sofar
    0.06
    .extra
    0.06
    	CHECK
    0.06
    Activations Density 0.009%

    No Known Activations