© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 2846
    Prev
    Next
    INDEX
    Explanations

    precise descriptive language or technical terms related to specific characteristics or qualities

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    909
    +0.08
    0.3%
    1446
    +0.08
    0.3%
    104
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1385
    +0.08
    0.15
    1499
    +0.08
    0.12
    104
    +0.06
    0.10
    Negative Logits
    <bos>
    -0.98
     intersper
    -0.88
     circums
    -0.73
     disambigu
    -0.70
     loggedIn
    -0.68
    ,
    -0.65
     Mlle
    -0.64
     punctu
    -0.64
    .
    -0.64
    -
    -0.63
    POSITIVE LOGITS
     utaf
    1.06
    MouseAdapter
    1.00
    yanto
    0.96
     Luglio
    0.93
     susun
    0.92
     carrefour
    0.91
     Février
    0.91
     Décembre
    0.91
     pymysql
    0.90
     alip
    0.90
    Activations Density 2.994%

    No Known Activations