Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 10325
    Prev
    Next
    INDEX
    Explanations

    references to specific conditions and characteristics within a broader context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AnchorStyles
    -0.51
    RemoveField
    -0.44
     lisäksi
    -0.44
    totalSupply
    -0.43
     innehå
    -0.43
    rungsseite
    -0.42
     nakalista
    -0.42
    gesamt
    -0.42
    dengar
    -0.42
     Ceux
    -0.42
    POSITIVE LOGITS
     region
    0.46
     regions
    0.46
     different
    0.45
    randomUUID
    0.43
     certain
    0.41
     différente
    0.41
     endforeach
    0.40
     territory
    0.40
     counties
    0.39
     geographic
    0.39
    Activations Density 0.679%

    No Known Activations