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-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 3088
    Prev
    Next
    INDEX
    Explanations

    references to detailed information, investigations, and historical research

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1150
    +0.07
    0.2%
    623
    +0.07
    0.2%
    924
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    193
    +0.07
    0.04
    1293
    +0.07
    0.04
    16
    +0.07
    0.05
    Negative Logits
     mépris
    -0.70
    
    
    -0.70
     ingrat
    -0.69
     getAge
    -0.68
     unspeak
    -0.65
     quitted
    -0.63
     monstre
    -0.62
     marchand
    -0.60
     malheur
    -0.60
     pamph
    -0.60
    POSITIVE LOGITS
    @[+][
    0.69
     <<<<<<<<<<<<<<
    0.63
    oneofs
    0.53
     vald
    0.52
    IntoConstraints
    0.51
    Initializable
    0.51
    IANGLES
    0.51
     дописавши
    0.50
    Viitteet
    0.49
    mtrl
    0.48
    Activations Density 0.402%

    No Known Activations