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-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 1706
    Prev
    Next
    INDEX
    Explanations

    references to financial figures and costs

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1334
    +0.18
    0.6%
    674
    +0.17
    0.6%
    1896
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1334
    +0.18
    0.13
    920
    +0.17
    0.07
    161
    +0.12
    0.07
    Negative Logits
    antity
    -0.76
    haviour
    -0.70
    esterday
    -0.69
    cember
    -0.68
    phalt
    -0.68
    intenance
    -0.66
     „,
    -0.66
    uestions
    -0.65
    iented
    -0.65
     Cæsar
    -0.65
    POSITIVE LOGITS
     Ainda
    0.60
    achieve
    0.59
     ensure
    0.59
     prevent
    0.58
     avoid
    0.56
    wieś
    0.56
     achieve
    0.55
     Talvez
    0.54
     comply
    0.54
     facilitate
    0.54
    Activations Density 0.527%

    No Known Activations