© 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. 10659
    Prev
    Next
    INDEX
    Explanations

    specific details related to a legal case involving accusations of overbilling for services

    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
    1553
    +0.10
    0.3%
    604
    +0.09
    0.3%
    1531
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1553
    +0.10
    0.07
    1531
    +0.09
    0.05
    804
    +0.09
    0.04
    Negative Logits
    <bos>
    -0.85
     parlar
    -0.64
    StoreMessageInfo
    -0.62
    endaft
    -0.59
     Trabal
    -0.55
    ieso
    -0.54
     seksi
    -0.53
     proprement
    -0.53
     morire
    -0.53
     Paling
    -0.52
    POSITIVE LOGITS
     unnecessarily
    0.92
     unnecessary
    0.83
     needlessly
    0.79
     incorrectly
    0.78
     wrong
    0.77
     inappropriate
    0.72
     unintended
    0.72
     wrongly
    0.72
     prematurely
    0.71
     incorrect
    0.70
    Activations Density 0.551%

    No Known Activations