© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 9185
    Prev
    Next
    INDEX
    Explanations

    references to bias and political influence in news reporting

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ëĬ¥
    -0.16
    yb
    -0.15
    licas
    -0.15
    _WRONG
    -0.14
    dff
    -0.13
    olini
    -0.13
    rey
    -0.13
    qa
    -0.13
    atile
    -0.13
    BadRequest
    -0.13
    POSITIVE LOGITS
     bias
    0.67
     Bias
    0.59
     biased
    0.59
     biases
    0.57
    Bias
    0.53
    bias
    0.52
    biased
    0.44
    _bias
    0.44
    åģı
    0.40
    .bias
    0.38
    Activations Density 0.061%

    No Known Activations