© 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. 8-RES-JB
    11. 297
    Prev
    Next
    INDEX
    Explanations

    the word "forget" followed by a concept or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "forget" and related phrases about forgetting or dismissing topics

    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
    emale
    -0.93
    æѦ
    -0.80
    similar
    -0.76
    osponsors
    -0.75
    osen
    -0.74
    WIND
    -0.71
    amen
    -0.70
    CVE
    -0.69
    éļ
    -0.69
    hiba
    -0.69
    POSITIVE LOGITS
     excuses
    0.86
     frivolous
    0.84
     fancy
    0.80
     pesky
    0.79
     kidding
    0.78
     worries
    0.77
     outright
    0.76
     frivol
    0.76
     distractions
    0.75
     gimm
    0.74
    Activations Density 0.129%

    No Known Activations