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. 1263
    Prev
    Next
    INDEX
    Explanations

    technical terms related to computer hardware and software

    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
    752
    +0.22
    0.7%
    1403
    +0.19
    0.6%
    2019
    +0.18
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    752
    +0.22
    0.05
    924
    +0.19
    0.05
    2019
    +0.18
    0.05
    Negative Logits
    oubted
    -0.88
    ecuted
    -0.83
    ertion
    -0.82
     epithe
    -0.82
     tubules
    -0.80
     tetrach
    -0.79
     setCode
    -0.78
     tubercle
    -0.77
     adenocarcinoma
    -0.77
    oubtedly
    -0.74
    POSITIVE LOGITS
     confé
    1.61
     catég
    1.37
     clô
    1.37
     vété
    1.36
     prétend
    1.34
     délib
    1.32
     véhic
    1.32
     gouver
    1.32
     perpétu
    1.30
     proprement
    1.27
    Activations Density 0.313%

    No Known Activations