© 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. 8-RES-JB
    11. 17707
    Prev
    Next
    INDEX
    Explanations

    future tense verbs indicating plans or intentions

    oai_token-act-pair · gpt-3.5-turbo

    conditional phrases and expressions of uncertainty regarding future actions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dictates
    -0.68
     arises
    -0.64
     indicates
    -0.64
     backdrop
    -0.64
     refres
    -0.62
     obsc
    -0.60
     Shots
    -0.60
     Insp
    -0.60
    earch
    -0.59
     Bravo
    -0.59
    POSITIVE LOGITS
     lose
    0.97
     get
    0.83
     succeed
    0.78
     someday
    0.78
     offend
    0.77
     starve
    0.77
    lost
    0.77
     hear
    0.77
     stumble
    0.76
     regain
    0.76
    Activations Density 0.342%

    No Known Activations