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. 11625
    Prev
    Next
    INDEX
    Explanations

    characters from a specific language or character set

    oai_token-act-pair · gpt-3.5-turbo

    certain special characters or symbols, particularly the character 'æ'

    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
    anwhile
    -0.97
    enegger
    -0.85
    espie
    -0.80
     Protector
    -0.76
     sclerosis
    -0.74
    rawdownloadcloneembedreportprint
    -0.72
    nyder
    -0.71
     Syndicate
    -0.69
     Keane
    -0.68
     proxies
    -0.67
    POSITIVE LOGITS
    ĻĤ
    1.45
    Ķ
    1.41
    İ
    1.37
    ¥µ
    1.35
    Ļ
    1.35
    Ĥª
    1.35
    ²
    1.33
    Ł
    1.31
    Ĭ
    1.31
    Ľ
    1.30
    Activations Density 0.005%

    No Known Activations