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. 10305
    Prev
    Next
    INDEX
    Explanations

    expressions of indifference and negligence, particularly in the context of social or moral responsibility

    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
     tarko
    -0.51
     myö
    -0.49
     dipa
    -0.48
     secretos
    -0.48
    IBOutlet
    -0.47
     köt
    -0.47
    hæng
    -0.46
    uvu
    -0.46
     jorden
    -0.45
     danni
    -0.45
    POSITIVE LOGITS
     indifference
    0.93
     indifferent
    0.90
    orance
    0.89
     AssemblyCulture
    0.77
     indiffer
    0.77
     inaction
    0.76
    __':
    
    0.73
     disregard
    0.73
     nonchal
    0.72
     ignor
    0.71
    Activations Density 0.495%

    No Known Activations