© 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. 0-RES-JB
    11. 1995
    Prev
    Next
    INDEX
    Explanations

    websites URLs containing the word "how"

    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.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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1871
    +0.09
    0.3%
    602
    +0.06
    0.2%
    1343
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.09
    0.04
    1343
    +0.06
    0.07
    1871
    +0.06
    0.03
    Negative Logits
    <bos>
    -1.18
    iddhar
    -0.81
     be
    -0.67
     przeds
    -0.67
     is
    -0.66
     then
    -0.66
     its
    -0.66
     has
    -0.65
     impelled
    -0.65
     will
    -0.65
    POSITIVE LOGITS
     kram
    2.08
     lele
    2.07
     ananas
    2.05
     meis
    1.92
     marte
    1.90
     Kategor
    1.87
     antik
    1.86
     uhr
    1.81
     elek
    1.81
     sement
    1.80
    Activations Density 0.318%

    No Known Activations