© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 19494
    Prev
    Next
    INDEX
    Explanations

    keywords related to engaging with or consuming content, such as "check," "read," "watch," "follow," and "join."

    oai_token-act-pair · gpt-3.5-turbo

    directives for reader engagement and interaction

    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
    Lenin
    -0.74
     guiActiveUnfocused
    -0.71
    BuyableInstoreAndOnline
    -0.71
     Loaded
    -0.65
     Hume
    -0.64
    Lens
    -0.64
     Hes
    -0.63
     Fell
    -0.62
     Lag
    -0.61
    rium
    -0.60
    POSITIVE LOGITS
     yourself
    1.03
     yourselves
    0.96
    ivably
    0.96
    veland
    0.77
    neau
    0.77
     easily
    0.73
    anchester
    0.72
     safely
    0.72
    uate
    0.72
    SOURCE
    0.72
    Activations Density 0.237%

    No Known Activations