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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 8-RES-JB
    11. 3324
    Prev
    Next
    INDEX
    Explanations

    terms related to cancellations and dismissals of various events or actions

    oai_token-act-pair · gpt-3.5-turbo

    terms related to cancellations, postponements, and halts in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Fit
    -0.75
    orers
    -0.71
    rics
    -0.67
    orians
    -0.64
    idth
    -0.64
    orio
    -0.63
    retch
    -0.62
    RIC
    -0.60
    Strength
    -0.59
    ses
    -0.59
    POSITIVE LOGITS
     indefinitely
    0.96
     by
    0.94
     altogether
    0.90
     unanimously
    0.86
     owing
    0.84
     due
    0.84
    uled
    0.83
     entirely
    0.82
     abruptly
    0.81
     because
    0.79
    Activations Density 0.166%

    No Known Activations