© 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. 11231
    Prev
    Next
    INDEX
    Explanations

    phrases related to news and trustworthy information

    oai_token-act-pair · gpt-3.5-turbo

    inquiries about reliable news sources

    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
    naires
    -0.63
    naire
    -0.62
    [/
    -0.59
     sucker
    -0.55
     fe
    -0.53
     Democr
    -0.53
     subconscious
    -0.52
     phys
    -0.51
    iter
    -0.51
     Azerb
    -0.51
    POSITIVE LOGITS
    isSpecialOrderable
    0.81
    soType
    0.66
    hig
    0.64
     ILCS
    0.64
    oother
    0.62
    channelAvailability
    0.60
    worthy
    0.60
    OME
    0.58
    maxwell
    0.57
    independence
    0.57
    Activations Density 0.032%

    No Known Activations