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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 14788
    Prev
    Next
    INDEX
    Explanations

    names of individuals and their relationships

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns, particularly names and places in familial contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     airspace
    -0.79
    asive
    -0.74
    Against
    -0.71
    safe
    -0.70
    CONCLUS
    -0.67
    cedented
    -0.67
     visualization
    -0.67
    Conclusion
    -0.67
     safer
    -0.67
     estimation
    -0.66
    POSITIVE LOGITS
    lda
    0.98
     Sr
    0.92
    aceae
    0.90
     grandfather
    0.89
     father
    0.86
    iblings
    0.84
     paternal
    0.83
     dynasty
    0.83
     Earl
    0.82
     family
    0.81
    Activations Density 0.250%

    No Known Activations