© 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. 4-RES-JB
    6. 4417
    Prev
    Next
    INDEX
    Explanations

    specific numbers or numerical ranges

    oai_token-act-pair · gpt-3.5-turbo

    numerical data, particularly related to percentages and confidence intervals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    soever
    -0.73
    roma
    -0.70
    cles
    -0.69
    pressure
    -0.68
    ged
    -0.68
    seeing
    -0.67
    âĹ¼
    -0.67
    ging
    -0.67
     Skydragon
    -0.67
    houses
    -0.67
    POSITIVE LOGITS
    %:
    0.95
    odcast
    0.80
     kHz
    0.76
    uppet
    0.74
     ILCS
    0.73
    isher
    0.73
    90
    0.72
    aneously
    0.71
    reme
    0.70
    99
    0.69
    Activations Density 0.019%

    No Known Activations