Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 34
    Prev
    Next
    INDEX
    Explanations

    phrases related to finance, specifically in the context of pets and wine regions

    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
    169
    +0.11
    0.3%
    757
    +0.10
    0.3%
    406
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1363
    +0.11
    0.02
    1120
    +0.10
    0.02
    406
    +0.09
    0.01
    Negative Logits
     intersper
    -0.69
    Dziękuję
    -0.61
    /*!
    
    -0.58
     sympathize
    -0.56
    pushd
    -0.53
     endeavored
    -0.53
     Óscar
    -0.51
     imprison
    -0.51
    César
    -0.51
     idolat
    -0.50
    POSITIVE LOGITS
     ::
    2.48
    ::
    2.42
    :::
    1.53
    ::$
    1.45
    ::_
    1.44
    >::
    1.42
     :::
    1.34
    (::
    1.32
    ::~
    1.25
    ::::
    1.24
    Activations Density 0.226%

    No Known Activations