© 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. 14916
    Prev
    Next
    INDEX
    Explanations

    information related to updates in a specific game or program

    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
    1577
    +0.45
    2.0%
    1842
    +0.28
    1.2%
    394
    +0.25
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.45
    0.10
    1577
    +0.28
    0.16
    1842
    +0.25
    0.12
    Negative Logits
     shenan
    -1.88
     encomp
    -1.85
     impra
    -1.82
     unspeak
    -1.80
     reluct
    -1.80
     affor
    -1.79
     indestru
    -1.78
     maneu
    -1.71
     increa
    -1.70
     scrat
    -1.69
    POSITIVE LOGITS
    MessageTagHelper
    0.58
    )).
    0.54
    .
    0.53
    WriteTagHelper
    0.52
    ‌.
    0.52
     
    0.52
    }}}{\
    0.52
     fieldNum
    0.51
    bibfnamefont
    0.51
     BoxFit
    0.51
    Activations Density 2.787%

    No Known Activations