© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 24-LLAMASCOPE-RES-32K
    11. 17329
    Prev
    Next
    INDEX
    Explanations

    abbreviated forms of words or phrases related to entities and actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    paring
    -0.28
    AndFeel
    -0.26
    nesday
    -0.26
    icipants
    -0.24
    iams
    -0.24
    auty
    -0.23
    bidden
    -0.23
    thing
    -0.23
    thern
    -0.23
    ificates
    -0.22
    POSITIVE LOGITS
    adays
    0.34
    odore
    0.30
    etheless
    0.29
    raries
    0.25
    alyze
    0.23
    anmar
    0.23
    atomy
    0.22
    aurants
    0.22
    pared
    0.20
    stitute
    0.20
    Activations Density 0.535%

    No Known Activations