© 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 EvalsExportsinterp-engineNEWAPI 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. 16353
    Prev
    Next
    INDEX
    Explanations

    phrases indicating an instruction, suggestion, or recommendation directed towards the reader, such as starting with "If you" and including a verb

    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
    1919
    +0.14
    0.4%
    381
    +0.13
    0.4%
    1978
    +0.12
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1919
    +0.14
    0.07
    1415
    +0.13
    0.04
    101
    +0.12
    0.04
    Negative Logits
     poliester
    -0.95
     poliuret
    -0.92
     moza
    -0.91
     utop
    -0.85
     parlamento
    -0.85
     répon
    -0.83
     vinil
    -0.83
     hunde
    -0.79
     archivio
    -0.79
     monaster
    -0.78
    POSITIVE LOGITS
     indestru
    1.10
     disagre
    1.06
     disreg
    1.02
     apprehen
    1.02
     impra
    1.00
     shenan
    1.00
     intersper
    0.97
     depic
    0.96
     unspeak
    0.96
     indescri
    0.95
    Activations Density 0.126%

    No Known Activations