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. 15770
    Prev
    Next
    INDEX
    Explanations

    unicode-related information and formats

    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
    957
    +0.08
    0.2%
    1343
    +0.08
    0.2%
    453
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.08
    0.04
    453
    +0.08
    0.05
    801
    +0.07
    0.04
    Negative Logits
    Zaj
    -0.53
    LookAnd
    -0.53
     Silurian
    -0.47
     صوتيه
    -0.46
    xtext
    -0.46
    flink
    -0.46
    transQ
    -0.46
     Wikimedijinoj
    -0.46
     barked
    -0.45
    Tó
    -0.45
    POSITIVE LOGITS
     soggior
    0.98
     cammin
    0.96
     Luglio
    0.94
     Ottobre
    0.92
     Settembre
    0.90
     Giugno
    0.89
     credere
    0.87
     insegna
    0.84
     domani
    0.83
     morire
    0.82
    Activations Density 0.210%

    No Known Activations