© 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. 6516
    Prev
    Next
    INDEX
    Explanations

    dates or numbers expressed in quantity or frequency

    oai_token-act-pair · gpt-3.5-turbo

    punctuated statements or exclamations

    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
    ¬¼
    -0.54
    OME
    -0.54
    sylv
    -0.53
    obin
    -0.52
    arius
    -0.51
    afety
    -0.50
    herent
    -0.49
    Ĩ
    -0.48
    rique
    -0.48
    alogue
    -0.48
    POSITIVE LOGITS
     etc
    0.96
     albeit
    0.67
    etc
    0.65
     respectively
    0.60
    76561
    0.58
     aka
    0.58
    pmwiki
    0.58
     Kinnikuman
    0.56
    atche
    0.51
     alas
    0.51
    Activations Density 0.694%

    No Known Activations