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

    Neuronpedia

    APICircuit 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. 650
    Prev
    Next
    INDEX
    Explanations

    instances of evidence suggesting documentary resources, possibly related to educational materials

    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
    1222
    +0.17
    0.9%
    32
    +0.15
    0.8%
    1335
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1222
    +0.17
    0.03
    1335
    +0.15
    0.03
    765
    +0.12
    0.03
    Negative Logits
    <bos>
    -2.31
    AsUp
    -0.68
    EndProject
    -0.63
     localize
    -0.63
    TargetException
    -0.62
    Obrázky
    -0.57
    RTLD
    -0.57
     synthesize
    -0.56
     materialize
    -0.56
    elemField
    -0.56
    POSITIVE LOGITS
     Pen
    1.45
    Pen
    1.40
     pen
    1.35
     PEN
    1.33
     Pens
    1.28
     pens
    1.25
    pen
    1.13
    Pens
    1.12
    PEN
    1.12
     péné
    1.11
    Activations Density 0.200%

    No Known Activations