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. 11-RES-JB
    6. 18232
    Prev
    Next
    INDEX
    Explanations

    legal and investigative terms

    oai_token-act-pair · gpt-3.5-turbo

    phrases associated with legal proceedings or formal inquiries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     crammed
    -0.93
     enchanted
    -0.91
     trendy
    -0.90
     invent
    -0.87
     dazzling
    -0.84
     fashionable
    -0.84
     hooked
    -0.84
     electr
    -0.83
     flashy
    -0.83
     fancy
    -0.83
    POSITIVE LOGITS
    Regarding
    1.52
    Statement
    1.43
    UPDATE
    1.40
    Therefore
    1.39
    Update
    1.34
    Thank
    1.34
    Furthermore
    1.34
    Please
    1.33
    Additionally
    1.32
    Secondly
    1.27
    Activations Density 0.515%

    No Known Activations