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-2-2B
    4. Residual Stream - 16k
    5. 4-GEMMASCOPE-RES-16K
    6. 3374
    Prev
    Next
    INDEX
    Explanations

    construction elements related to programming or data handling, particularly focusing on object-oriented and structured data types

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SequentialGroup
    -1.15
    httphttps
    -1.03
     nakalista
    -1.01
     onAnimation
    -0.89
    AndEndTag
    -0.86
    IntoConstraints
    -0.84
    Personendaten
    -0.82
    Примітки
    -0.82
     rospy
    -0.81
    abstractmethod
    -0.79
    POSITIVE LOGITS
     pel
    0.70
     Ak
    0.59
     Pel
    0.59
     amb
    0.53
     So
    0.52
     "
    0.52
     di
    0.51
     I
    0.50
    EndOf
    0.48
     inner
    0.48
    Activations Density 0.051%

    No Known Activations