© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 3578
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences discussing or emphasizing the veracity or accuracy of statements or claims

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    the word "true" and sometimes nearby words indicating agreement or confirmation.

    oai_token-act-pair · gemini-1.5-flashTriggered by @shantanudarveshi
    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
    1127
    +0.15
    0.5%
    406
    +0.14
    0.5%
    605
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1127
    +0.15
    0.04
    406
    +0.14
    0.03
    1370
    +0.11
    0.03
    Negative Logits
    )_/¯
    -0.55
    MLLoader
    -0.54
    Ảnh
    -0.47
    serviceWorker
    -0.46
    ModelAdmin
    -0.45
    gruntled
    -0.44
     paraf
    -0.43
     wireType
    -0.43
     onAnimation
    -0.42
    HPP
    -0.42
    POSITIVE LOGITS
     chrétien
    0.76
    stdbool
    0.72
     embodi
    0.71
     truer
    0.70
     true
    0.68
     reconno
    0.68
     Keny
    0.68
     intersper
    0.67
    True
    0.65
     untrue
    0.65
    Activations Density 0.055%

    No Known Activations