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. 15369
    Prev
    Next
    INDEX
    Explanations

    positive evaluations and comparisons

    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
    1257
    +0.09
    0.2%
    62
    +0.08
    0.2%
    12
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    725
    +0.09
    0.04
    1358
    +0.08
    0.07
    62
    +0.07
    0.05
    Negative Logits
    ⇐
    -0.63
     manuten
    -0.61
     hunde
    -0.61
     praktik
    -0.61
     keramik
    -0.60
     prega
    -0.60
     Bakter
    -0.60
     adal
    -0.59
     akus
    -0.59
     alkoh
    -0.58
    POSITIVE LOGITS
     but
    1.51
    but
    1.29
     nhưng
    1.24
     But
    1.20
    But
    1.18
     BUT
    1.18
     pero
    1.10
     แต่
    1.01
    BUT
    1.00
     lakini
    1.00
    Activations Density 1.056%

    No Known Activations