© 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. 1361
    Prev
    Next
    INDEX
    Explanations

    words related to the city of Pittsburgh within sports context

    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
    1708
    +0.12
    0.4%
    789
    +0.09
    0.3%
    172
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    789
    +0.12
    0.03
    1590
    +0.09
    0.03
    1708
    +0.09
    0.03
    Negative Logits
    setDo
    -0.76
     continue
    -0.75
     ensure
    -0.74
     add
    -0.72
     enable
    -0.71
    var
    -0.70
     fix
    -0.70
     adapt
    -0.70
     survive
    -0.70
     carry
    -0.70
    POSITIVE LOGITS
     ftu
    2.16
     desir
    2.11
     fuf
    2.09
     wien
    2.01
     fta
    2.01
     madonna
    2.00
     eiffel
    1.99
     applau
    1.99
     fup
    1.97
     perfet
    1.94
    Activations Density 0.109%

    No Known Activations