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. 0-RES-JB
    6. 21794
    Prev
    Next
    INDEX
    Explanations

    phrases or titles with the word "Miss" followed by a person's name or a specific context

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Miss" in various contexts

    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
    endi
    -0.73
    URI
    -0.67
     awoken
    -0.66
    ãĥĩ
    -0.65
    éĥ
    -0.64
    tools
    -0.64
     bin
    -0.64
    ual
    -0.63
     centralized
    -0.63
    bas
    -0.62
    POSITIVE LOGITS
     Miss
    4.10
    Miss
    2.97
     MISS
    2.32
    miss
    2.16
     Mrs
    1.45
     miss
    1.45
     pageant
    1.43
    Mrs
    1.38
     Ms
    1.36
     Mistress
    1.28
    Activations Density 0.023%

    No Known Activations