© 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. 1070
    Prev
    Next
    INDEX
    Explanations

    words related to punk culture, such as 'punk', 'hardcore punk', and references to punk bands

    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
    82
    +0.11
    0.3%
    1350
    +0.11
    0.3%
    442
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1363
    +0.11
    0.03
    690
    +0.11
    0.03
    1343
    +0.09
    0.03
    Negative Logits
    makeText
    -0.67
     pö
    -0.56
     nõ
    -0.56
     pü
    -0.56
     effekt
    -0.56
     spørgsmål
    -0.55
     maksi
    -0.54
     всичко
    -0.54
    Према
    -0.53
    ph
    -0.53
    POSITIVE LOGITS
     Punk
    2.47
     punk
    2.46
    Punk
    2.34
    punk
    2.19
     fluo
    1.40
     Cæ
    1.37
     increa
    1.34
     embodi
    1.27
     indestru
    1.26
     maneu
    1.24
    Activations Density 0.223%

    No Known Activations