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. 8253
    Prev
    Next
    INDEX
    Explanations

    numbers in a specific format: a decimal point followed by one or two digits

    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
    1343
    +0.20
    0.6%
    674
    +0.15
    0.5%
    382
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    382
    +0.20
    0.04
    1607
    +0.15
    0.03
    1429
    +0.13
    0.03
    Negative Logits
     shenan
    -1.30
     unspeak
    -1.24
     intersper
    -1.20
     reluct
    -1.16
     maneu
    -1.16
     impra
    -1.13
     indescri
    -1.08
     disreg
    -1.08
     horrend
    -1.08
     unavoid
    -1.06
    POSITIVE LOGITS
     habet
    0.96
     marte
    0.94
     alkoh
    0.91
     potest
    0.89
     utop
    0.88
     ananas
    0.85
     solidar
    0.85
     manuten
    0.83
     pól
    0.83
     bambu
    0.82
    Activations Density 0.052%

    No Known Activations