© 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. 5106
    Prev
    Next
    INDEX
    Explanations

    descriptions of violent physical altercations or assaults

    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
    +0.13
    0.4%
    946
    +0.12
    0.4%
    50
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    946
    +0.13
    0.20
    100
    +0.12
    0.11
    970
    +0.12
    0.09
    Negative Logits
     susun
    -0.70
     trouva
    -0.70
     eorum
    -0.69
     veu
    -0.67
     potes
    -0.64
     aspetta
    -0.64
     habet
    -0.63
     quæ
    -0.63
     CiNii
    -0.62
     tomba
    -0.61
    POSITIVE LOGITS
     Bartholo
    1.06
     Bengt
    1.01
     Knud
    0.96
     McLaugh
    0.93
     Sted
    0.91
     Oester
    0.88
     Olof
    0.86
     reluct
    0.86
     unspeak
    0.85
     McInt
    0.82
    Activations Density 5.601%

    No Known Activations