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. 0-RES-JB
    6. 11132
    Prev
    Next
    INDEX
    Explanations

    Russian verbs in their infinitive form

    oai_token-act-pair · gpt-3.5-turbo

    specific characters or symbols from a non-English language

    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
     GT
    -0.74
     clock
    -0.72
     waist
    -0.69
     Riley
    -0.67
     Temple
    -0.64
     Giles
    -0.63
     gag
    -0.63
     Clover
    -0.61
     shrine
    -0.61
     sensitivity
    -0.61
    POSITIVE LOGITS
    в
    4.72
    м
    2.42
    к
    2.35
    д
    2.34
    н
    2.02
    л
    2.01
    ÑĤ
    1.90
    ÑĢ
    1.89
    Ð
    1.83
    ÑĮ
    1.68
    Activations Density 0.011%

    No Known Activations