© 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 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. 25-LLAMASCOPE-RES-32K
    11. 12722
    Prev
    Next
    INDEX
    Explanations

    phrases indicating intentions or plans related to future actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    expressions of intent or necessity, particularly phrases with modal verbs like "need to," "have to," "going to," "might have to," and "can."

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    statements expressing planned, potential, or obligated actions, typically using modal verbs with “to” infinitive constructions.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    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
    Å¡ov
    -0.18
    nech
    -0.15
    abin
    -0.15
    ÑĢеб
    -0.15
    ASURE
    -0.14
    ç·ł
    -0.14
    izzly
    -0.14
    fds
    -0.14
     Kobe
    -0.13
    _Ptr
    -0.13
    POSITIVE LOGITS
     figure
    0.19
    blog
    0.19
     finish
    0.16
     finally
    0.16
    figure
    0.16
     fig
    0.16
     proper
    0.15
     finishing
    0.15
     properly
    0.15
    Finish
    0.15
    Activations Density 0.167%

    No Known Activations