© 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. 12786
    Prev
    Next
    INDEX
    Explanations

    punctuation marks like periods and quotation marks

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and sentence-ending markers, particularly periods

    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
     depl
    -0.92
     unsus
    -0.84
     indis
    -0.82
     teasp
    -0.82
     conspic
    -0.81
     repud
    -0.80
    Þ
    -0.77
     unamb
    -0.76
     adequately
    -0.76
     grievance
    -0.75
    POSITIVE LOGITS
     Sometimes
    1.52
     Then
    1.49
     Usually
    1.44
     Especially
    1.43
     Luckily
    1.42
     Eventually
    1.42
     Obviously
    1.41
     But
    1.37
     Hopefully
    1.36
     Whereas
    1.35
    Activations Density 0.282%

    No Known Activations