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. 15605
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical objects like tools and weapon names

    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
    1385
    +0.10
    0.3%
    1220
    +0.08
    0.2%
    1363
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1220
    +0.10
    0.04
    1819
    +0.08
    0.05
    736
    +0.08
    0.04
    Negative Logits
     alkoh
    -1.13
     kram
    -1.07
     lü
    -1.06
     kras
    -1.03
     maksi
    -1.01
     uhr
    -1.00
     panik
    -1.00
     minimalis
    -0.98
     augus
    -0.96
     hek
    -0.96
    POSITIVE LOGITS
     wielded
    0.67
     resultList
    0.66
     OBSERVATIONS
    0.62
     ryzen
    0.61
     resultMap
    0.61
     earnestness
    0.59
     holdt
    0.59
     sticks
    0.58
     APPARATUS
    0.58
     itemList
    0.58
    Activations Density 0.302%

    No Known Activations