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. 0-RES-JB
    6. 19429
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "anguish."

    oai_token-act-pair · gpt-3.5-turbo

    words related to language or linguistic themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     parody
    -0.70
     mega
    -0.67
     digit
    -0.67
    Jones
    -0.65
     MSG
    -0.64
     Jenny
    -0.63
     til
    -0.63
     Merry
    -0.63
     ms
    -0.62
     Katz
    -0.61
    POSITIVE LOGITS
    angu
    4.64
    ANG
    1.73
    ingu
    1.41
    antry
    1.09
    oub
    1.07
    anguard
    1.04
    ang
    1.03
    agu
    1.02
    avage
    1.01
    anqu
    1.00
    Activations Density 0.030%

    No Known Activations