© 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 EvalsExportsinterp-engineNEWAPI 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. 44692
    Prev
    Next
    INDEX
    Explanations

    references to specific details or notable elements in a narrative

    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
    izr
    -0.15
    /gcc
    -0.15
    FillColor
    -0.15
    اض
    -0.15
    -en
    -0.15
    ichert
    -0.14
    ToProps
    -0.14
     Clo
    -0.14
    EObject
    -0.14
    ousel
    -0.13
    POSITIVE LOGITS
     catch
    0.69
     caught
    0.68
     catches
    0.66
     catching
    0.61
    catch
    0.59
     Catch
    0.56
    caught
    0.54
    Catch
    0.54
     Caught
    0.49
    Caught
    0.46
    Activations Density 0.063%

    No Known Activations