© 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. 17226
    Prev
    Next
    INDEX
    Explanations

    statements expressing feelings of violation or distress related to personal rights or safety

    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
     sort
    -0.20
     terrific
    -0.16
    sort
    -0.15
     pars
    -0.15
    -redux
    -0.15
     sorts
    -0.15
     interesting
    -0.14
     fewer
    -0.14
    unes
    -0.14
    acer
    -0.14
    POSITIVE LOGITS
    691
    0.15
     sir
    0.15
     Freem
    0.15
    ionario
    0.15
    .obtain
    0.14
     ÙħتÙĨ
    0.14
    _vect
    0.14
    vasion
    0.14
     terminology
    0.14
     Sir
    0.14
    Activations Density 0.072%

    No Known Activations