Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 30-LLAMASCOPE-RES-32K
    6. 23396
    Prev
    Next
    INDEX
    Explanations

    instances of the word "break" and its various forms

    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
    eid
    -0.16
    amet
    -0.15
    opoulos
    -0.15
    phy
    -0.15
    irth
    -0.15
    دÙĪ
    -0.15
    aura
    -0.15
    ikan
    -0.14
    imos
    -0.14
    rr
    -0.14
    POSITIVE LOGITS
    fast
    0.36
    age
    0.34
    away
    0.33
    neck
    0.32
    water
    0.28
    FAST
    0.28
    ages
    0.27
    dance
    0.25
    through
    0.24
    aways
    0.24
    Activations Density 0.040%

    No Known Activations