© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 2469
    Prev
    Next
    INDEX
    Explanations

    specific numerical values related to a quantity or statistic

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the number 32 and its variations in different 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
    soDeliveryDate
    -0.81
     è£ıè
    -0.64
    aura
    -0.63
    é»Ĵ
    -0.63
    amaru
    -0.60
    witch
    -0.59
    alth
    -0.58
    ãĥĬ
    -0.58
     metab
    -0.57
    tm
    -0.57
    POSITIVE LOGITS
     32
    3.04
     64
    2.28
     33
    2.20
     34
    2.11
     31
    2.03
     36
    1.98
     42
    1.93
     38
    1.90
     37
    1.87
     62
    1.85
    Activations Density 0.023%

    No Known Activations