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. 15516
    Prev
    Next
    INDEX
    Explanations

    identification information such as enrollment IDs, numbers, and details associated with deceased individuals

    oai_token-act-pair · gpt-3.5-turbo

    terms related to identification and enrollment processes, particularly involving a government ID system

    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
    storms
    -0.84
    ovie
    -0.82
     layoffs
    -0.72
    rawdownloadcloneembedreportprint
    -0.71
     turbulence
    -0.71
     starve
    -0.71
     worsen
    -0.70
     explosions
    -0.70
     improv
    -0.69
     intensify
    -0.69
    POSITIVE LOGITS
     passport
    1.27
     authentication
    1.25
     fingerprint
    1.21
     password
    1.18
     passports
    1.16
     identification
    1.16
    password
    1.16
     verification
    1.11
     IDs
    1.09
     registration
    1.09
    Activations Density 0.661%

    No Known Activations