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. 8-RES-JB
    6. 23163
    Prev
    Next
    INDEX
    Explanations

    emotionally charged expressions and symbols

    oai_token-act-pair · gpt-3.5-turbo

    uncommon or special characters used in digital contexts

    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
     Gillespie
    -0.78
     gloom
    -0.69
     Sidd
    -0.68
     Franch
    -0.67
    ioned
    -0.65
     Sind
    -0.63
     Quadro
    -0.63
     Sle
    -0.63
     Abyssal
    -0.63
     Tid
    -0.62
    POSITIVE LOGITS
    İ
    1.60
    Ĵ
    1.60
    Į
    1.57
    ĵ
    1.55
    Ķ
    1.48
    ĻĤ
    1.46
    ¥ŀ
    1.43
    į
    1.40
    IJ
    1.38
    ¦
    1.37
    Activations Density 0.012%

    No Known Activations