© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 4191
    Prev
    Next
    INDEX
    Explanations

    references to father figures

    oai_token-act-pair · gpt-3.5-turbo

    references to familial roles, particularly focusing on the terms "father" and "mother."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     reality
    -0.73
     limited
    -0.69
     cup
    -0.68
     procedural
    -0.66
     clutter
    -0.66
     restricted
    -0.65
     secure
    -0.65
     ha
    -0.65
     run
    -0.65
     arena
    -0.64
    POSITIVE LOGITS
    father
    3.84
    mother
    3.21
    parents
    2.39
    daughter
    1.93
    brother
    1.91
    Father
    1.82
    children
    1.67
    dad
    1.49
    Mother
    1.45
    child
    1.38
    Activations Density 0.023%

    No Known Activations