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. 4-RES-JB
    6. 5670
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names containing the string "uch"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "uch" or similar sequences of sounds

    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
    mith
    -0.68
    xtap
    -0.67
    ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
    -0.67
     eaves
    -0.66
     Logged
    -0.64
    hig
    -0.63
     Quit
    -0.63
    ences
    -0.63
    ©¶æ
    -0.62
    vernment
    -0.61
    POSITIVE LOGITS
    icago
    1.00
    anan
    0.92
    lord
    0.71
    arist
    0.71
    reon
    0.70
     Parenthood
    0.70
    uch
    0.70
    hhhh
    0.69
    tm
    0.68
    atan
    0.67
    Activations Density 0.007%

    No Known Activations