© 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. 0-RES-JB
    11. 4218
    Prev
    Next
    INDEX
    Explanations

    the word "forthcoming."

    oai_token-act-pair · gpt-3.5-turbo

    terms related to upcoming events or releases

    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
    umber
    -0.79
    mare
    -0.64
    ias
    -0.61
     Ruler
    -0.60
    OOOO
    -0.59
     vs
    -0.58
    olor
    -0.57
     sided
    -0.57
    addle
    -0.55
     instead
    -0.55
    POSITIVE LOGITS
     forthcoming
    4.01
     upcoming
    1.43
    forth
    1.21
     imminent
    1.11
     elusive
    1.03
     impending
    1.02
     hereafter
    0.99
     pending
    0.96
     extant
    0.90
     candid
    0.87
    Activations Density 0.013%

    No Known Activations