© 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. 7607
    Prev
    Next
    INDEX
    Explanations

    terms related to laws, regulations, and online terms of service

    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
    2019
    +0.12
    0.4%
    1415
    +0.11
    0.3%
    1510
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1510
    +0.12
    0.05
    1265
    +0.11
    0.04
    299
    +0.09
    0.04
    Negative Logits
     espé
    -1.23
     monaster
    -1.20
     fré
    -1.20
     utop
    -1.18
     alkoh
    -1.17
     silikon
    -1.12
     éta
    -1.12
     hunde
    -1.11
     asfal
    -1.11
     sement
    -1.09
    POSITIVE LOGITS
     participate
    0.91
     create
    0.89
     utilize
    0.87
     investigate
    0.86
     perform
    0.86
     engage
    0.85
     undertake
    0.84
     develop
    0.84
     interact
    0.84
     pursue
    0.84
    Activations Density 0.405%

    No Known Activations