© 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. 25-LLAMASCOPE-RES-32K
    11. 10334
    Prev
    Next
    INDEX
    Explanations

    phrases indicating abandonment or neglect

    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
    lav
    -0.17
    ÏĥÏĦÏĮ
    -0.16
    wang
    -0.16
    avra
    -0.15
     disguise
    -0.15
    kel
    -0.14
    akh
    -0.14
    ,$_
    -0.14
    GU
    -0.13
     Keeping
    -0.13
    POSITIVE LOGITS
     behind
    0.40
     alone
    0.37
     Behind
    0.33
    alone
    0.32
    Behind
    0.30
    beh
    0.28
     Alone
    0.28
    -alone
    0.25
     aside
    0.24
    à¹Ħว
    0.22
    Activations Density 0.052%

    No Known Activations