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. 13951
    Prev
    Next
    INDEX
    Explanations

    specialized terminology and expressions related to technical subjects

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

    starting with "ex" or "und"

    np_acts-logits-general · gemini-2.0-flash

    ex, und, eld prefixes

    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
    er
    -2.97
    ER
    -1.58
    erà
    -1.41
    erate
    -1.39
    erii
    -1.22
    erl
    -1.19
    erent
    -1.12
    erai
    -1.10
    erus
    -1.09
    erle
    -1.09
    POSITIVE LOGITS
    ERS
    0.63
     degradability
    0.54
    logical
    0.54
    erex
    0.52
    ʒ
    0.51
    hoid
    0.51
     paramInt
    0.49
    bewerken
    0.49
    erv
    0.48
     tartalomajánló
    0.48
    Activations Density 0.386%

    No Known Activations