© 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 SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 6-RES-JB
    11. 3920
    Prev
    Next
    INDEX
    Explanations

    email-related phrases like "send link" or "send email"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.20
    1.2%
    1178
    +0.10
    0.6%
    849
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    849
    +0.20
    0.05
    1451
    +0.10
    0.04
    991
    +0.10
    0.04
    Negative Logits
    <bos>
    -3.42
    /*
    -0.80
    struct
    -0.73
    ⓧ
    -0.72
    hyrchwyd
    -0.72
    HasAnnotation
    -0.71
    <>
    -0.68
    <?
    -0.67
    itemize
    -0.66
    //}
    -0.66
    POSITIVE LOGITS
     increa
    2.11
     impra
    2.08
     affor
    2.07
     reluct
    2.07
     maneu
    2.02
     emphat
    2.01
     disagre
    2.00
     unden
    1.99
     accla
    1.98
     indestru
    1.98
    Activations Density 0.204%

    No Known Activations