© 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. 4-RES-JB
    11. 15964
    Prev
    Next
    INDEX
    Explanations

    phrases related to data analysis, research patterns, and trends

    oai_token-act-pair · gpt-3.5-turbo

    references to patterns and trends in various contexts

    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
    ned
    -0.84
    INGTON
    -0.73
    ded
    -0.73
    ESA
    -0.69
    ×ŀ
    -0.68
    gha
    -0.67
    ille
    -0.67
    MENT
    -0.66
    oÄŁ
    -0.64
    clamation
    -0.63
    POSITIVE LOGITS
    etting
    1.23
    uggest
    1.10
    hops
    1.10
    omething
    1.05
    pace
    1.03
    ettings
    1.02
    etter
    1.00
    cape
    1.00
    poons
    0.98
    hips
    0.98
    Activations Density 0.057%

    No Known Activations