© 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. 11-RES-JB
    11. 7523
    Prev
    Next
    INDEX
    Explanations

    negations or statements indicating absence or lack of something

    oai_token-act-pair · gpt-3.5-turbo

    negative phrases or terms related to performance or capability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DragonMagazine
    -0.92
    agate
    -0.75
    OPLE
    -0.69
    onomy
    -0.65
    )=(
    -0.64
    renheit
    -0.61
     compan
    -0.60
    iox
    -0.60
    FIG
    -0.60
    TeX
    -0.59
    POSITIVE LOGITS
    ched
    1.17
    ching
    1.12
     score
    0.90
     scoring
    0.87
    ranked
    0.85
     postseason
    0.84
     disappoint
    0.84
    scoring
    0.83
     flashy
    0.82
     scored
    0.79
    Activations Density 0.174%

    No Known Activations