© 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. 30-LLAMASCOPE-RES-32K
    11. 22854
    Prev
    Next
    INDEX
    Explanations

    elements related to font styles and sizes in CSS code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ors
    -0.19
    ooter
    -0.18
    leck
    -0.17
    umont
    -0.17
    ofire
    -0.16
    phant
    -0.16
    orsch
    -0.16
    thic
    -0.16
    766
    -0.16
    iyah
    -0.16
    POSITIVE LOGITS
    -family
    0.33
    -face
    0.31
    -weight
    0.28
     family
    0.28
     face
    0.27
    face
    0.27
    Face
    0.27
    awesome
    0.27
    _family
    0.25
    _HERSHEY
    0.25
    Activations Density 0.004%

    No Known Activations