© 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. 22261
    Prev
    Next
    INDEX
    Explanations

    instances where something could have been avoided

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "avoided" and related concepts

    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
    wow
    -0.71
    eem
    -0.60
     dyn
    -0.59
     vest
    -0.58
     Shard
    -0.57
     grant
    -0.55
    plex
    -0.55
    morph
    -0.55
     premiere
    -0.54
     mail
    -0.54
    POSITIVE LOGITS
     avoided
    3.63
     dodged
    2.13
     avoids
    2.02
     aver
    1.88
     avoiding
    1.71
    oided
    1.68
    avoid
    1.68
     avoid
    1.62
     skipped
    1.55
     prevented
    1.51
    Activations Density 0.013%

    No Known Activations