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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. MLP Out - 131k
    9. 6-GEMMASCOPE-MLP-131K
    11. 110040
    Prev
    Next
    INDEX
    Explanations

    contractions and possessive forms related to actions and states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.6.hook_mlp_out
    Hook Layer
    6
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     resourceCulture
    -0.60
     betweenstory
    -0.57
    AddTagHelper
    -0.56
     protoimpl
    -0.56
    AccessorTable
    -0.53
     <=",
    -0.53
    exels
    -0.51
    ContentAsync
    -0.50
    +#+#
    -0.47
     שוליים
    -0.46
    POSITIVE LOGITS
     nė
    0.43
    fortawesome
    0.41
    previous
    0.40
    //
    0.39
     vorige
    0.38
    ⁦
    0.38
     Bahnhof
    0.37
     Constitucional
    0.36
    brinco
    0.36
    IConfiguration
    0.36
    Activations Density 0.241%

    No Known Activations