© 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. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 15-LLAMASCOPE-RES-32K
    11. 19191
    Prev
    Next
    INDEX
    Explanations

    references to deception and impersonation in criminal contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    isay
    -0.20
    SPATH
    -0.15
    weg
    -0.15
    AdapterManager
    -0.15
    combe
    -0.15
    iquer
    -0.14
    _PCI
    -0.14
    íĹĮ
    -0.14
    quest
    -0.14
    PCI
    -0.13
    POSITIVE LOGITS
     convinc
    0.23
     convincing
    0.22
     fake
    0.18
    è£Ŀ
    0.17
    /fa
    0.17
     Fake
    0.16
    overn
    0.16
     plausible
    0.16
     believable
    0.16
    fake
    0.15
    Activations Density 0.075%

    No Known Activations