© 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. 22115
    Prev
    Next
    INDEX
    Explanations

    phrases related to controversial issues and falsehoods

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and phrases that indicate relationships or 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
    Mart
    -0.90
    arms
    -0.89
    AMS
    -0.89
    chnology
    -0.89
    arm
    -0.89
    enegger
    -0.88
    ername
    -0.88
    acon
    -0.86
    amer
    -0.85
    ARM
    -0.84
    POSITIVE LOGITS
     flo
    1.12
     spl
    1.07
     bleed
    1.05
     dup
    1.01
     Spl
    0.99
     drip
    0.97
    Spl
    0.97
     plung
    0.96
     plunge
    0.93
     dro
    0.93
    Activations Density 0.417%

    No Known Activations