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. 273
    Prev
    Next
    INDEX
    Explanations

    titles or roles related to various professions and organizations

    oai_token-act-pair · gpt-3.5-turbo

    titles or roles associated with individuals, particularly in professional 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
    zers
    -0.73
    cards
    -0.72
     informants
    -0.71
    worldly
    -0.71
    ¬¼
    -0.68
     pieces
    -0.66
    goers
    -0.65
    oner
    -0.65
    pieces
    -0.65
     recordings
    -0.65
    POSITIVE LOGITS
     MPH
    0.91
     Sierra
    0.85
     NCT
    0.82
     USAF
    0.81
     Kinnikuman
    0.80
     Ministry
    0.80
     NEC
    0.78
     Carnegie
    0.75
     Dept
    0.74
     Programming
    0.74
    Activations Density 0.132%

    No Known Activations