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. 9-RES-JB
    6. 21088
    Prev
    Next
    INDEX
    Explanations

    words related to spice or flavor

    oai_token-act-pair · gpt-3.5-turbo

    sequences of text starting with the "<|endoftext|>" token and occurrences of the character 'k'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    single token corresponding to ' k'

    Explanation Uploaded by User

    word that starts with ' k'

    Explanation Uploaded by User
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •words starting with k

    No Comments

    Negative Logits
     Malf
    -0.69
    Ø©
    -0.66
     chronological
    -0.66
    terday
    -0.65
     Murdoch
    -0.64
     TTL
    -0.64
     expires
    -0.63
     unlaw
    -0.62
     Staples
    -0.62
     galleries
    -0.59
    POSITIVE LOGITS
    etchup
    1.31
    itty
    1.30
    idd
    1.23
    appa
    1.19
    ratom
    1.18
    ansas
    1.18
    udos
    1.15
    orea
    1.14
    rieg
    1.11
    idding
    1.08
    Activations Density 0.033%

    No Known Activations