© 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. 14645
    Prev
    Next
    INDEX
    Explanations

    dates in history

    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
    1363
    +0.10
    0.3%
    1870
    +0.09
    0.3%
    1741
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1614
    +0.10
    0.04
    841
    +0.09
    0.03
    1368
    +0.09
    0.03
    Negative Logits
     unintelligible
    -0.57
     conjectures
    -0.56
     philosophic
    -0.56
     vicissitudes
    -0.54
     monotone
    -0.53
     miscon
    -0.52
     artifice
    -0.52
     sophistic
    -0.51
     disambigu
    -0.51
     resemb
    -0.51
    POSITIVE LOGITS
     moza
    1.18
     alkoh
    1.17
     karton
    1.15
     parati
    1.15
     maksi
    1.14
     traktor
    1.13
     elek
    1.10
     keramik
    1.10
     kask
    1.09
     makro
    1.09
    Activations Density 0.051%

    No Known Activations