Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 376
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names related to specific events or locations, maybe in the context of news or announcements

    oai_token-act-pair · gpt-3.5-turbo

    word patterns associated with sounds or rhythm

    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
     Fathers
    -0.92
    Winged
    -0.78
     menstru
    -0.73
    uct
    -0.70
     Companion
    -0.67
    Sync
    -0.66
    alcohol
    -0.65
    Gl
    -0.64
     Darkness
    -0.64
    Gallery
    -0.64
    POSITIVE LOGITS
    OUNT
    2.26
    OUN
    2.07
    ONG
    1.75
    isal
    1.51
    pun
    1.37
    ilon
    1.29
     racket
    1.28
    bourg
    1.24
    atson
    1.20
    bnb
    1.10
    Activations Density 0.048%

    No Known Activations