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. 5618
    Prev
    Next
    INDEX
    Explanations

    references and mentions of insight or understanding in various contexts

    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
    s
    -0.99
     queſta
    -0.53
    theless
    -0.48
    stanbul
    -0.48
    ities
    -0.47
     getY
    -0.46
    iiiiiiii
    -0.45
    いきます
    -0.45
    SenderId
    -0.44
    IIIIIIII
    -0.44
    POSITIVE LOGITS
    <bos>
    0.79
    𝓮
    0.66
    𝓵
    0.65
    𝓭
    0.63
    𝓹
    0.62
    𝓲
    0.62
    𝓫
    0.61
    𝓴
    0.61
    ítmény
    0.60
    റ്
    0.58
    Activations Density 1.072%

    No Known Activations