Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 2254
    Prev
    Next
    INDEX
    Explanations

    special characters commonly found in non-English languages

    oai_token-act-pair · gpt-3.5-turbo

    special characters and symbols, possibly indicating formatting or coding structures

    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
     mathemat
    -0.79
    rero
    -0.79
    milo
    -0.78
     recourse
    -0.78
     advant
    -0.75
     contrace
    -0.74
     compr
    -0.73
     behav
    -0.72
     tradem
    -0.72
    ategory
    -0.71
    POSITIVE LOGITS
    ·
    1.25
    в
    1.13
    а
    1.08
    ÏĤ
    1.07
    м
    1.05
    ´
    0.99
    Ñĭ
    0.99
    é¾į
    0.96
    ׾
    0.95
    ¸
    0.94
    Activations Density 0.007%

    No Known Activations