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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11296
    Prev
    Next
    INDEX
    Explanations

    phrases indicating importance or attention to specific details

    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
    1793
    +0.12
    0.4%
    872
    +0.10
    0.3%
    460
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1793
    +0.12
    0.05
    1601
    +0.10
    0.03
    890
    +0.08
    0.03
    Negative Logits
    <bos>
    -0.52
     abbra
    -0.52
    Seni
    -0.51
    WithIdentifier
    -0.51
     monaster
    -0.49
     Demok
    -0.49
     legislat
    -0.48
     Himmels
    -0.48
     lega
    -0.48
    Viungo
    -0.47
    POSITIVE LOGITS
     unspeak
    0.95
     disagre
    0.95
     hairc
    0.91
     unwarran
    0.91
     Ename
    0.88
     ecru
    0.86
     swee
    0.86
     affor
    0.82
     tolerably
    0.81
     gaily
    0.80
    Activations Density 0.139%

    No Known Activations