© 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
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 821
    Prev
    Next
    INDEX
    Explanations

    software features and functions related to computer security and device protection

    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
    50
    +0.12
    0.5%
    1194
    +0.06
    0.3%
    1964
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1194
    +0.12
    0.05
    380
    +0.06
    0.04
    647
    +0.06
    0.04
    Negative Logits
    <bos>
    -1.12
     regain
    -0.69
    ⓧ
    -0.69
     shivered
    -0.63
     deepen
    -0.63
    public
    -0.63
     shuddered
    -0.62
     eradicate
    -0.62
     reunite
    -0.62
     continue
    -0.61
    POSITIVE LOGITS
     Pro
    2.03
    Pro
    1.72
     véhic
    1.54
     quoique
    1.52
     renfer
    1.42
     quelquefois
    1.41
     obé
    1.40
     rafra
    1.40
     soulign
    1.37
     prétend
    1.37
    Activations Density 0.117%

    No Known Activations