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. 4-RES-JB
    6. 23474
    Prev
    Next
    INDEX
    Explanations

    mentions of ages under 18 or groups identified by their age-range

    oai_token-act-pair · gpt-3.5-turbo

    references to age-specific categories or groups

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

    No Comments

    Negative Logits
    è¦ļéĨĴ
    -0.83
    dest
    -0.77
    ãĤ¯
    -0.68
    Effects
    -0.67
    Dest
    -0.64
    916
    -0.63
    output
    -0.63
    finding
    -0.63
    ciating
    -0.63
    atl
    -0.62
    POSITIVE LOGITS
    enegger
    0.80
     century
    0.76
    apist
    0.74
     era
    0.70
     Franch
    0.68
    erer
    0.67
    foundland
    0.67
    yright
    0.66
     Franchise
    0.66
     Pinball
    0.66
    Activations Density 0.331%

    No Known Activations