© 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. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 29-LLAMASCOPE-RES-32K
    11. 19898
    Prev
    Next
    INDEX
    Explanations

    instances of the word "refuse" or its variations indicating a lack of compliance or agreement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.29.hook_resid_post
    Hook Layer
    29
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ctl
    -0.15
    clusive
    -0.15
    ago
    -0.15
    cha
    -0.15
    agem
    -0.15
     Reputation
    -0.14
    loub
    -0.14
    berapa
    -0.14
    igan
    -0.14
    aggio
    -0.14
    POSITIVE LOGITS
     any
    0.31
     requests
    0.31
     offers
    0.25
     pleas
    0.24
    requests
    0.23
     invitations
    0.23
     altogether
    0.23
     demands
    0.23
     calls
    0.22
    ä»»ä½ķ
    0.21
    Activations Density 0.059%

    No Known Activations