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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 3733
    Prev
    Next
    INDEX
    Explanations

    numbers indicating temporal sequences

    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
    776
    +0.16
    0.5%
    200
    +0.14
    0.5%
    897
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    200
    +0.16
    0.05
    474
    +0.14
    0.04
    517
    +0.13
    0.03
    Negative Logits
     Secrétaire
    -0.63
     Puoi
    -0.61
     Græ
    -0.61
     Hæ
    -0.60
     autunno
    -0.59
     saar
    -0.58
     Eccle
    -0.58
     Traité
    -0.57
    alnız
    -0.57
     soste
    -0.57
    POSITIVE LOGITS
     last
    1.06
     LAST
    1.01
    last
    0.96
     Last
    0.94
    Last
    0.89
    LAST
    0.89
    setLast
    0.81
    getLast
    0.76
     getLast
    0.74
    ultimo
    0.66
    Activations Density 0.087%

    No Known Activations