© 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. 10561
    Prev
    Next
    INDEX
    Explanations

    specific phrases indicating capability or achievement

    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
    +0.18
    0.6%
    2019
    +0.15
    0.5%
    1334
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1415
    +0.18
    0.06
    1334
    +0.15
    0.05
    869
    +0.13
    0.04
    Negative Logits
     chré
    -1.00
     Meksi
    -0.98
     matel
    -0.97
     quelquefois
    -0.96
     Româ
    -0.96
     quoique
    -0.95
     obé
    -0.94
     désol
    -0.93
     bourgeo
    -0.92
     véhic
    -0.92
    POSITIVE LOGITS
     achieve
    0.72
     accommodate
    0.68
     safely
    0.68
     obtain
    0.65
    ഊ
    0.63
     afford
    0.62
     successfully
    0.61
    Kontra
    0.61
     generate
    0.61
     withstand
    0.61
    Activations Density 0.160%

    No Known Activations