© 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. 4117
    Prev
    Next
    INDEX
    Explanations

    cancellations or actions related to canceling something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Cancel."

    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
    reader
    -0.76
    III
    -0.76
    velt
    -0.75
     NRS
    -0.72
    geist
    -0.71
    Lex
    -0.66
    Spirit
    -0.66
    read
    -0.66
    DX
    -0.65
    zzy
    -0.65
    POSITIVE LOGITS
    ides
    0.75
     bargaining
    0.72
     withd
    0.68
    */(
    0.66
    icum
    0.64
     submarines
    0.64
    uitous
    0.63
    angan
    0.63
     pione
    0.62
     verbs
    0.62
    Activations Density 0.000%

    No Known Activations