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. 24-GEMMASCOPE-RES-16K
    6. 4502
    Prev
    Next
    INDEX
    Explanations

    the preposition "in" and its various contexts within the text

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

    "in agreement" or "in line"

    np_acts-logits-general · gemini-2.0-flash

    in agreement, in line, in accordance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    setVerticalGroup
    -0.66
     lèvres
    -0.58
     antemano
    -0.58
     Autorisations
    -0.57
    llion
    -0.56
     חיצוניים
    -0.54
     calendriers
    -0.53
     læng
    -0.53
     étoient
    -0.52
    koľvek
    -0.52
    POSITIVE LOGITS
     keeping
    1.06
     association
    0.99
    keeping
    0.98
     alignment
    0.97
     correlation
    0.97
     line
    0.95
     accordance
    0.94
     collaboration
    0.93
     response
    0.92
     conjunction
    0.89
    Activations Density 0.420%

    No Known Activations