© 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. 11-LLAMASCOPE-RES-32K
    11. 3393
    Prev
    Next
    INDEX
    Explanations

    statements about capabilities or states of being, particularly related to actions and effects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     comm
    -0.16
    úa
    -0.15
    iker
    -0.15
     empir
    -0.14
    ifty
    -0.14
    STANCE
    -0.14
    osphere
    -0.14
     POSSIBILITY
    -0.14
    STREAM
    -0.14
    ikers
    -0.13
    POSITIVE LOGITS
    rous
    0.16
    ahr
    0.15
    è°
    0.15
    hil
    0.15
    hod
    0.14
     Baum
    0.14
    quin
    0.14
    λικ
    0.14
    ORM
    0.14
    rod
    0.14
    Activations Density 0.662%

    No Known Activations