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. 22-GEMMASCOPE-RES-16K
    6. 8251
    Prev
    Next
    INDEX
    Explanations

    instances of temporal phrases indicating the passage of time

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

    "Since" followed by a variety of words

    np_acts-logits-general · gemini-2.0-flash

    since a point in time

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    zehn
    -0.57
    บัติ
    -0.56
    canActivate
    -0.49
    یی
    -0.48
    outputFile
    -0.48
    erschied
    -0.47
    digan
    -0.46
    hecta
    -0.46
    oredCriteria
    -0.45
    es
    -0.44
    POSITIVE LOGITS
     inception
    1.02
     childhood
    0.89
     antiquity
    0.82
     then
    0.82
     صوتيه
    0.78
     day
    0.76
     arriving
    0.74
     então
    0.74
     Beginn
    0.72
    })`
    0.70
    Activations Density 0.065%

    No Known Activations