© 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. 4999
    Prev
    Next
    INDEX
    Explanations

    multi-word phrases that introduce additional points or arguments

    oai_token-act-pair · gpt-3.5-turbo

    references to sequencing or iterations of ideas or points

    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
    ULTS
    -1.02
    umers
    -0.92
    ographs
    -0.92
    endars
    -0.91
    ateurs
    -0.90
    ENTS
    -0.88
    osponsors
    -0.86
    ouls
    -0.84
    isons
    -0.83
    ampires
    -0.83
    POSITIVE LOGITS
     pecul
    1.53
     caveat
    1.48
     flaw
    1.42
     aspect
    1.40
     factor
    1.39
     misconception
    1.37
     phenomenon
    1.33
     complication
    1.32
     problem
    1.30
     reason
    1.28
    Activations Density 0.277%

    No Known Activations