© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 3883
    Prev
    Next
    INDEX
    Explanations

    components related to computer hardware, particularly storage devices like hard drives and SSDs

    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
    876
    +0.17
    0.5%
    1385
    +0.10
    0.3%
    1267
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    936
    +0.17
    0.03
    678
    +0.10
    0.03
    1443
    +0.09
    0.02
    Negative Logits
     pamph
    -0.76
     Moslem
    -0.69
     Silurian
    -0.68
     Rumania
    -0.62
     viciss
    -0.62
     Jä
    -0.61
     Keny
    -0.61
     Parli
    -0.61
     subjek
    -0.60
     Ruman
    -0.59
    POSITIVE LOGITS
     storage
    0.76
     Storage
    0.66
     drives
    0.65
     disk
    0.64
     drive
    0.60
     exé
    0.60
    storage
    0.60
     Drives
    0.59
     gmbh
    0.59
     tupperware
    0.59
    Activations Density 0.211%

    No Known Activations