© 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
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 13343
    Prev
    Next
    INDEX
    Explanations

    terms related to a specific concept, possibly involving lightning, seizure, epidemiology, genetic disorders, children, mountains, retailers, suppliers, and principles

    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
    1967
    +0.19
    0.6%
    1438
    +0.12
    0.4%
    1984
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1438
    +0.19
    0.07
    1984
    +0.12
    0.07
    1654
    +0.10
    0.06
    Negative Logits
    Autoritní
    -1.02
    OGND
    -0.86
     autorytatywna
    -0.84
    Bibliograf
    -0.83
    SharedDtor
    -0.83
     الاطلاع
    -0.82
    writeFieldEnd
    -0.81
     invokingState
    -0.80
     AssemblyCulture
    -0.79
     kaarangay
    -0.79
    POSITIVE LOGITS
     shenan
    1.23
     unspeak
    1.17
     apprehen
    1.15
     reluct
    1.09
     intersper
    1.06
     disagre
    1.06
     indescri
    1.04
     encomp
    1.03
     inappro
    1.00
     ineffec
    0.99
    Activations Density 1.098%

    No Known Activations