Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 6-RES-JB
    6. 5715
    Prev
    Next
    INDEX
    Explanations

    information regarding terms and conditions, privacy policies, and personal information protection in legal, contractual, and educational contexts

    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.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.29
    1.4%
    658
    +0.15
    0.7%
    478
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    658
    +0.29
    0.17
    1023
    +0.15
    0.13
    1937
    +0.13
    0.11
    Negative Logits
    <bos>
    -3.44
    ///**
    -0.66
    public
    -0.64
     which
    -0.64
    struct
    -0.61
    ,
    -0.60
    lateinit
    -0.58
     for
    -0.58
    //
    -0.58
     that
    -0.57
    POSITIVE LOGITS
     Minang
    1.41
     guarante
    1.35
     ftu
    1.32
     sappi
    1.31
     effe
    1.30
     reft
    1.28
     increa
    1.26
     stockholm
    1.26
     perfon
    1.26
     ftre
    1.25
    Activations Density 1.156%

    No Known Activations