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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 729
    Prev
    Next
    INDEX
    Explanations

    references to national parks or related events and services

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    63
    +0.07
    0.2%
    1783
    +0.07
    0.2%
    300
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.07
    0.02
    491
    +0.07
    0.02
    1343
    +0.06
    0.03
    Negative Logits
     intersper
    -1.62
     encomp
    -1.60
     depic
    -1.47
     fuf
    -1.42
     attemp
    -1.41
     shenan
    -1.40
     erad
    -1.40
     volunte
    -1.38
     reluct
    -1.38
     increa
    -1.36
    POSITIVE LOGITS
     tanong
    0.87
    StreetMap
    0.82
    cèse
    0.74
    GeoNames
    0.72
    كويكب
    0.71
     estimés
    0.70
    fillText
    0.70
    requipa
    0.69
    ilarang
    0.67
    monix
    0.66
    Activations Density 0.282%

    No Known Activations