© 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. 10-RES-JB
    11. 12449
    Prev
    Next
    INDEX
    Explanations

    phrases related to conducting investigations or inspections

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to official investigations or reviews

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    liners
    -0.74
    Joined
    -0.73
     behaved
    -0.72
     hurled
    -0.69
    why
    -0.68
     Virtue
    -0.67
    <?
    -0.67
    ovable
    -0.67
     preached
    -0.66
     trope
    -0.66
    POSITIVE LOGITS
     investigation
    1.38
     analysis
    1.38
     research
    1.37
     evaluation
    1.37
     tests
    1.36
     survey
    1.31
     inspections
    1.30
     surveys
    1.30
     feasibility
    1.30
     testing
    1.29
    Activations Density 0.335%

    No Known Activations