Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 11518
    Prev
    Next
    INDEX
    Explanations

    phrases related to awe-inspiring or impressive things or events

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "spectacular" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    appa
    -0.83
    ©¶æ
    -0.78
    claimer
    -0.77
    aternity
    -0.73
    REL
    -0.72
    iltr
    -0.69
    lessness
    -0.67
    unker
    -0.67
    ploma
    -0.65
    avis
    -0.65
    POSITIVE LOGITS
    ly
    1.17
    ulously
    0.90
    NESS
    0.87
    LY
    0.83
     feats
    0.82
     spectacular
    0.82
     spectacle
    0.79
    ity
    0.79
    uously
    0.78
     rendition
    0.77
    Activations Density 0.006%

    No Known Activations