© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 394
    Prev
    Next
    INDEX
    Explanations

    words related to specific people or topics, particularly focusing on football and possibly politics

    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
    893
    +0.13
    0.4%
    241
    +0.08
    0.3%
    805
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    893
    +0.13
    0.04
    1310
    +0.08
    0.03
    1343
    +0.07
    0.03
    Negative Logits
    addComponent
    -0.84
    Configurer
    -0.80
    addGroup
    -0.79
    Unmarshaller
    -0.79
    cout
    -0.77
    public
    -0.74
    EndContext
    -0.73
    echo
    -0.72
    GroupLayout
    -0.72
    <bos>
    -0.72
    POSITIVE LOGITS
     maneu
    2.47
     affor
    2.46
     increa
    2.44
     scrat
    2.31
     excru
    2.31
     shenan
    2.28
     indestru
    2.26
     accla
    2.23
     impra
    2.22
     reluct
    2.22
    Activations Density 0.194%

    No Known Activations