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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 916
    Prev
    Next
    INDEX
    Explanations

    mentions of geographical locations ending in 'ung'

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1492
    +0.09
    0.3%
    971
    +0.09
    0.3%
    1778
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1343
    +0.09
    0.04
    1363
    +0.09
    0.03
    1120
    +0.08
    0.03
    Negative Logits
     intersper
    -1.10
     unspeak
    -1.06
     shenan
    -1.00
     indescri
    -0.93
     exasper
    -0.92
     downvote
    -0.91
     downvotes
    -0.91
     Wtf
    -0.90
     encomp
    -0.90
     odious
    -0.88
    POSITIVE LOGITS
    ung
    2.75
    UNG
    2.04
     ung
    1.84
    Ung
    1.74
     Ung
    1.62
    ungs
    1.45
    ungen
    1.44
    unga
    1.15
    unge
    1.14
    tung
    0.97
    Activations Density 0.230%

    No Known Activations