© 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. 4718
    Prev
    Next
    INDEX
    Explanations

    comparisons or contrasts between different concepts or scenarios

    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
    +0.15
    0.4%
    1622
    +0.11
    0.3%
    1624
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1622
    +0.15
    0.03
    706
    +0.11
    0.03
    1307
    +0.10
    0.02
    Negative Logits
     lele
    -1.17
     uhr
    -1.15
     Ibidem
    -1.14
     Intere
    -1.14
     Teks
    -1.13
     disagre
    -1.12
     Sén
    -1.10
     maksi
    -1.09
     hcm
    -1.09
     keramik
    -1.07
    POSITIVE LOGITS
    <bos>
    1.11
     much
    0.63
     so
    0.58
     hard
    0.55
     great
    0.54
     easy
    0.53
    стма
    0.52
     super
    0.50
     quite
    0.50
    митрий
    0.50
    Activations Density 0.111%

    No Known Activations