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. 8423
    Prev
    Next
    INDEX
    Explanations

    references to historical texts and architectural elements

    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
    1871
    +0.14
    0.4%
    872
    +0.11
    0.3%
    509
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1188
    +0.14
    0.02
    924
    +0.11
    0.05
    237
    +0.11
    0.02
    Negative Logits
    <bos>
    -0.75
    quehanna
    -0.63
     CreateTagHelper
    -0.61
     smör
    -0.61
     Violon
    -0.61
     lapto
    -0.60
    rosion
    -0.60
    styleType
    -0.59
     frambo
    -0.58
    DoubleQuotes
    -0.57
    POSITIVE LOGITS
     clô
    0.90
     étend
    0.86
     renforcé
    0.86
     librement
    0.84
     élar
    0.83
     renfer
    0.82
     nettement
    0.82
     quoique
    0.82
     spécialement
    0.82
     doté
    0.81
    Activations Density 0.475%

    No Known Activations