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. 85947
    Prev
    Next
    INDEX
    Explanations

    Cryptographic Hash Functions

    np_max-act · gemini-2.0-flash

    The neuron is detecting occurrences of the “MD5” hash algorithm name (e.g. the literal “MD5” token).

    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
    _elem
    -0.07
    效
    -0.07
    .tree
    -0.07
     QUAL
    -0.07
     URLWithString
    -0.07
     кус
    -0.06
     eagerly
    -0.06
     کرده
    -0.06
     долж
    -0.06
     đa
    -0.06
    POSITIVE LOGITS
    ового
    0.06
    bc
    0.06
    xxx
    0.06
    Mutable
    0.06
     misleading
    0.06
    +offset
    0.06
    BERT
    0.06
    _FM
    0.06
    .Question
    0.06
    .firstname
    0.05
    Activations Density 0.008%

    No Known Activations