© 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. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 1050
    Prev
    Next
    INDEX
    Explanations

    potential negative societal implications, particularly related to politics and economics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    674
    +0.39
    2.7%
    1253
    +0.06
    0.4%
    1741
    +0.06
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    800
    +0.39
    0.27
    354
    +0.06
    0.25
    1523
    +0.06
    0.24
    Negative Logits
     unspeak
    -4.39
     reluct
    -4.16
     shenan
    -4.09
     disagre
    -3.88
     apprehen
    -3.85
     impra
    -3.81
     indestru
    -3.67
     ineffec
    -3.65
     horrend
    -3.59
     disgra
    -3.58
    POSITIVE LOGITS
    <bos>
    13.56
    expandindo
    2.65
    GEBURTSDATUM
    2.61
     betweenstory
    2.61
    Autoritní
    2.58
     Paglinawan
    2.50
     Italijani
    2.38
     Walkover
    2.33
     '\\;'
    2.28
    تقاوى
    2.28
    Activations Density 0.081%

    No Known Activations