© 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. 5291
    Prev
    Next
    INDEX
    Explanations

    phrases related to organizations and awards

    oai_token-act-pair · gpt-3.5-turbo

    references to specific ratings or classifications related to quality metrics

    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
     Marx
    -0.77
     bowel
    -0.71
     vex
    -0.70
     congen
    -0.69
     chron
    -0.68
     ful
    -0.66
     Gord
    -0.65
    staking
    -0.65
     Prelude
    -0.65
     lunch
    -0.65
    POSITIVE LOGITS
    AA
    4.23
    AAA
    2.62
    AAAA
    1.87
    AAF
    1.80
     AA
    1.78
    aa
    1.56
    AAAAAAAA
    1.51
    ACA
    1.46
    UA
    1.44
    BA
    1.43
    Activations Density 0.011%

    No Known Activations