© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 14240
    Prev
    Next
    INDEX
    Explanations

    phrases related to providing assistance or support

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    The word "help" and nearby words in contexts expressing capability or purpose to achieve beneficial outcomes.

    oai_token-act-pair · claude-4-5-haikuTriggered by @areeb0102
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    think
    -0.43
    #+#
    -0.42
     go
    -0.41
     nakalista
    -0.41
    Ask
    -0.39
     think
    -0.37
    ask
    -0.36
     Ask
    -0.35
    speak
    -0.34
     Spend
    -0.34
    POSITIVE LOGITS
     enhance
    1.34
     improve
    1.32
     reduce
    1.26
     promote
    1.26
     prevent
    1.20
     maximize
    1.20
     protect
    1.16
     minimize
    1.15
     facilitate
    1.14
     simplify
    1.12
    Activations Density 0.846%

    No Known Activations