© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 13890
    Prev
    Next
    INDEX
    Explanations

    numeric values, likely focusing on numbers with patterns or specific ranges

    oai_token-act-pair · gpt-3.5-turbo

    numerical values, specifically referencing the number 47

    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
    urally
    -0.68
    inventoryQuantity
    -0.62
     organ
    -0.61
    ric
    -0.59
     Salt
    -0.59
     motion
    -0.58
    istic
    -0.58
     fibre
    -0.57
    llor
    -0.56
     filtering
    -0.56
    POSITIVE LOGITS
    47
    3.04
    48
    2.07
    46
    1.96
    43
    1.78
    49
    1.73
    51
    1.70
     47
    1.68
    45
    1.65
    53
    1.63
    67
    1.61
    Activations Density 0.043%

    No Known Activations