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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 2262
    Prev
    Next
    INDEX
    Explanations

    mentions of siblings, specifically older brothers

    oai_token-act-pair · gpt-3.5-turbo

    references to siblings, specifically "brother."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ifact
    -0.75
    issions
    -0.75
    orney
    -0.72
    aceous
    -0.71
    pmwiki
    -0.68
    veyard
    -0.68
    ibrary
    -0.68
    upuncture
    -0.67
    erity
    -0.66
    argon
    -0.65
    POSITIVE LOGITS
    hood
    1.46
     brother
    1.03
    heses
    0.98
    brother
    0.95
     brothers
    0.94
    hesis
    0.91
    pins
    0.88
    friend
    0.88
    Brother
    0.87
     patriarch
    0.79
    Activations Density 0.018%

    No Known Activations