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. 17824
    Prev
    Next
    INDEX
    Explanations

    numbers in a specific format (##)

    oai_token-act-pair · gpt-3.5-turbo

    instances of the number 34 in various contexts

    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
    •Embedding Space Digit Features

    No Comments

    Negative Logits
    etts
    -0.78
    ney
    -0.71
     Quan
    -0.66
    rington
    -0.61
    vern
    -0.60
     Savior
    -0.60
     Burns
    -0.60
     Wilson
    -0.59
    ewater
    -0.59
    ep
    -0.59
    POSITIVE LOGITS
    34
    2.27
     34
    2.16
     334
    1.46
     344
    1.44
     134
    1.41
     33
    1.41
    334
    1.40
    134
    1.40
     54
    1.35
    344
    1.31
    Activations Density 0.042%

    No Known Activations