© 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. 25-LLAMASCOPE-RES-32K
    11. 18313
    Prev
    Next
    INDEX
    Explanations

    references to the concept of balance in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    VOICE
    -0.16
    ollah
    -0.16
    sWith
    -0.15
    lum
    -0.15
    esting
    -0.15
    zin
    -0.15
    lings
    -0.15
    ling
    -0.15
    lage
    -0.15
    iel
    -0.15
    POSITIVE LOGITS
     sheet
    0.25
     Sheet
    0.21
    -sheet
    0.20
     sheets
    0.19
     between
    0.19
    Sheet
    0.18
     scales
    0.18
     prec
    0.17
    _sheet
    0.17
     beam
    0.17
    Activations Density 0.025%

    No Known Activations