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. 7-RES-JB
    6. 15958
    Prev
    Next
    INDEX
    Explanations

    phrases where someone is emphasizing a point or sharing surprising information

    oai_token-act-pair · gpt-3.5-turbo

    the rhetorical phrase "you know what," often used to emphasize a point or lead into a statement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aches
    -0.72
    still
    -0.63
    hammad
    -0.59
    robe
    -0.59
    ãģ®éŃĶ
    -0.58
    erate
    -0.58
    gem
    -0.58
    eer
    -0.58
    adden
    -0.57
    actor
    -0.56
    POSITIVE LOGITS
     happened
    1.08
     happ
    1.01
     happens
    1.00
     else
    0.91
     ensued
    0.81
     transpired
    0.72
    soever
    0.72
     separates
    0.71
    ciating
    0.69
    !?
    0.68
    Activations Density 0.060%

    No Known Activations