© 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. 20529
    Prev
    Next
    INDEX
    Explanations

    specific keywords that signal a new piece of information or a plot twist in a narrative

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Turn" and similar terms indicating transitions or actions

    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
     otherwise
    -0.65
     challeng
    -0.63
     reper
    -0.58
     peel
    -0.56
    AAF
    -0.55
    aimon
    -0.55
     Haram
    -0.55
     territ
    -0.54
    fortunately
    -0.53
    YS
    -0.53
    POSITIVE LOGITS
    ing
    2.38
    ed
    1.76
    edIn
    1.37
    ership
    1.27
    erness
    1.17
    edly
    1.13
    ingham
    1.10
    er
    1.10
    ING
    1.10
    ers
    1.06
    Activations Density 0.244%

    No Known Activations