© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 20296
    Prev
    Next
    INDEX
    Explanations

    phrases indicating comparison or quantity, particularly the word "more"

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate comparisons, quantifications, and reflections on social or economic conditions

    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
     teasp
    -0.61
    Sharp
    -0.59
    ĺħ
    -0.59
    ;;;;;;;;;;;;
    -0.55
     Again
    -0.54
    çļ
    -0.54
    AAAAAAAA
    -0.54
    }}}
    -0.54
    ãħĭ
    -0.52
     slightest
    -0.51
    POSITIVE LOGITS
     than
    2.53
    than
    2.47
     Than
    1.86
    Th
    0.83
     TH
    0.80
     compared
    0.75
    worldly
    0.75
     then
    0.71
    erous
    0.67
    then
    0.67
    Activations Density 0.643%

    No Known Activations