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. 8452
    Prev
    Next
    INDEX
    Explanations

    mentions of legal or financial terms related to bail

    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
    1092
    +0.17
    0.6%
    1895
    +0.14
    0.5%
    1622
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1092
    +0.17
    0.03
    1351
    +0.14
    0.01
    860
    +0.12
    0.02
    Negative Logits
     fei
    -0.94
     mme
    -0.85
     myn
    -0.82
     fars
    -0.81
     inder
    -0.80
     tanga
    -0.79
     marte
    -0.79
     franz
    -0.79
     kac
    -0.78
     unden
    -0.78
    POSITIVE LOGITS
     bail
    1.32
    bail
    1.11
    Bail
    1.04
     Bail
    0.99
     bailout
    0.87
     ba
    0.64
     baile
    0.56
    ypeł
    0.56
     linkovi
    0.55
    knitted
    0.55
    Activations Density 0.067%

    No Known Activations