© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 706
    Prev
    Next
    INDEX
    Explanations

    links or call-to-action buttons for further information or engagement on a digital platform

    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
    331
    +0.08
    0.2%
    1582
    +0.07
    0.2%
    1467
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1011
    +0.08
    0.03
    1582
    +0.07
    0.03
    1679
    +0.07
    0.02
    Negative Logits
     must
    -0.87
     continue
    -0.86
     don
    -0.86
     post
    -0.86
     go
    -0.85
     but
    -0.85
     get
    -0.85
     met
    -0.84
     so
    -0.83
     له
    -0.83
    POSITIVE LOGITS
     increa
    3.31
     stockholm
    3.11
     squa
    3.10
     disagre
    3.10
     reluct
    3.10
     guarante
    3.08
     secon
    3.07
     effe
    3.07
     inev
    3.02
     maneu
    3.02
    Activations Density 0.177%

    No Known Activations