© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 54257
    Prev
    Next
    INDEX
    Explanations

    criteria related to weight class limits and eligibility for competitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åij½
    -0.15
    _values
    -0.14
     VALUES
    -0.14
     Helmet
    -0.14
    -generic
    -0.13
    eneric
    -0.13
     Values
    -0.13
    pus
    -0.13
    itz
    -0.13
    obre
    -0.13
    POSITIVE LOGITS
     weight
    0.43
    weight
    0.36
     Weight
    0.33
     weights
    0.31
    Weight
    0.30
    .weight
    0.30
     WEIGHT
    0.29
    -weight
    0.29
     weigh
    0.29
    weights
    0.28
    Activations Density 0.015%

    No Known Activations