© 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. 7695
    Prev
    Next
    INDEX
    Explanations

    phrases describing visual actions like looking or staring

    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
    906
    +0.11
    0.3%
    1533
    +0.08
    0.2%
    446
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    753
    +0.11
    0.01
    240
    +0.08
    0.02
    1490
    +0.08
    0.03
    Negative Logits
     bandung
    -0.99
     jaya
    -0.96
     fernando
    -0.91
     kasa
    -0.91
     haup
    -0.90
     milano
    -0.89
     felipe
    -0.89
     ricardo
    -0.88
     soggior
    -0.88
     istan
    -0.87
    POSITIVE LOGITS
     eyes
    0.57
    ământ
    0.51
     resourceCulture
    0.51
    FieldDescriptor
    0.50
    eyes
    0.50
     looked
    0.49
     CreateTagHelper
    0.49
    Personendaten
    0.48
    atific
    0.48
     propOrder
    0.47
    Activations Density 0.156%

    No Known Activations