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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 86
    Prev
    Next
    INDEX
    Explanations

    references to government security agreements and historical figures

    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
    690
    +0.10
    0.4%
    752
    +0.09
    0.4%
    1871
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1080
    +0.10
    0.10
    823
    +0.09
    0.10
    1871
    +0.09
    0.03
    Negative Logits
     do
    -0.97
     continue
    -0.94
     become
    -0.93
     also
    -0.92
     ensure
    -0.92
     come
    -0.92
     don
    -0.92
    public
    -0.92
    Ngoài
    -0.90
     have
    -0.89
    POSITIVE LOGITS
     vété
    2.74
     stockholm
    2.61
     Traité
    2.54
     Chapitre
    2.48
     confé
    2.45
     véhic
    2.41
     Juf
    2.40
     clô
    2.38
     soulign
    2.38
     délib
    2.36
    Activations Density 1.070%

    No Known Activations