© 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. 68
    Prev
    Next
    INDEX
    Explanations

    words related to geographical locations and potential future developments or projects

    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
    832
    +0.09
    0.2%
    206
    +0.07
    0.2%
    1108
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.09
    0.02
    283
    +0.07
    0.02
    304
    +0.07
    0.02
    Negative Logits
     on
    -0.70
    </strong>
    -0.70
     in
    -0.69
     at
    -0.67
    </b>
    -0.66
     he
    -0.65
     by
    -0.65
     to
    -0.64
     through
    -0.64
     de
    -0.63
    POSITIVE LOGITS
     dises
    1.85
     haup
    1.76
     hcm
    1.74
     jaya
    1.74
     sappi
    1.73
     hina
    1.68
     napoli
    1.61
     santiago
    1.59
     umo
    1.58
     seiz
    1.58
    Activations Density 0.311%

    No Known Activations