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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 6250
    Prev
    Next
    INDEX
    Explanations

    expressions of struggle and conflict, particularly in the context of battles against personal or societal challenges

    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
    ]")]
    -0.63
     ***!
    -0.60
     queſto
    -0.57
    <pad>
    -0.55
     dAtA
    -0.54
    urator
    -0.54
     Picchu
    -0.54
    <unused43>
    -0.54
    <unused41>
    -0.54
    <unused23>
    -0.53
    POSITIVE LOGITS
     fight
    1.05
     Fight
    1.01
    fight
    1.00
     FIGHT
    1.00
    Fight
    0.98
     fought
    0.96
    FIGHT
    0.90
     fighting
    0.90
     battle
    0.90
    Fighting
    0.87
    Activations Density 0.259%

    No Known Activations