© 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. 9-RES-JB
    11. 8715
    Prev
    Next
    INDEX
    Explanations

    contractions of negations, such as "don't" or "doesn't"

    oai_token-act-pair · gpt-3.5-turbo

    negations or phrases that indicate absence or denial

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çĶŁ
    -0.71
    PDATE
    -0.68
    itiz
    -0.68
     Laun
    -0.66
    jected
    -0.65
     Opportun
    -0.65
     è£ıè
    -0.65
    stre
    -0.64
     unforeseen
    -0.64
    estern
    -0.63
    POSITIVE LOGITS
     necessarily
    1.28
     seem
    1.23
     belong
    1.14
     deserve
    1.12
     condone
    1.12
     discriminate
    1.10
     exist
    1.09
     intend
    1.09
     appear
    1.07
     hesitate
    1.04
    Activations Density 0.125%

    No Known Activations