© 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. 814
    Prev
    Next
    INDEX
    Explanations

    negative words and critical statements

    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
    946
    +0.11
    0.3%
    596
    +0.08
    0.2%
    1253
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    716
    +0.11
    0.02
    71
    +0.08
    0.02
    1990
    +0.07
    0.03
    Negative Logits
     cuck
    -0.51
     wif
    -0.45
     accla
    -0.45
     nobly
    -0.43
    syms
    -0.42
     Kien
    -0.42
     Daven
    -0.42
     Bartholo
    -0.42
     diametro
    -0.41
    strokeStyle
    -0.41
    POSITIVE LOGITS
    /**
    0.70
     slightest
    0.61
     sker
    0.59
     nė
    0.57
     julk
    0.57
     začet
    0.56
    lloworld
    0.54
     praça
    0.52
     conç
    0.52
     sentiers
    0.51
    Activations Density 0.187%

    No Known Activations