© 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. 10-RES-JB
    11. 7431
    Prev
    Next
    INDEX
    Explanations

    negative statements or uncertainties

    oai_token-act-pair · gpt-3.5-turbo

    negative contractions and phrases suggesting a lack of success or fulfillment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kamp
    -0.68
    ãģ®ç
    -0.65
     arsen
    -0.62
    ologically
    -0.58
    Reviewer
    -0.57
     appropri
    -0.57
    udic
    -0.56
    itech
    -0.55
    agate
    -0.55
     administr
    -0.54
    POSITIVE LOGITS
     relent
    0.79
     dissu
    0.77
    stop
    0.73
     dull
    0.70
     appe
    0.68
     smoothly
    0.66
     deter
    0.66
     smooth
    0.65
     quiet
    0.65
    aired
    0.65
    Activations Density 0.216%

    No Known Activations