© 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. 22440
    Prev
    Next
    INDEX
    Explanations

    proper nouns starting with "Par" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Par" and its variations, suggesting a focus on references or discussions involving paragraphs or sections of text

    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
     endorsement
    -0.75
     Blumenthal
    -0.73
     toll
    -0.68
     Braz
    -0.67
     AE
    -0.66
     Cu
    -0.66
     Weiss
    -0.65
     CF
    -0.63
     brown
    -0.63
     mug
    -0.63
    POSITIVE LOGITS
    Par
    3.64
    PAR
    1.66
    par
    1.64
     Par
    1.58
    Pan
    1.58
    Param
    1.35
     par
    1.30
    Parser
    1.23
    params
    1.22
    Pand
    1.22
    Activations Density 0.010%

    No Known Activations