© 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. 8-RES-JB
    11. 17725
    Prev
    Next
    INDEX
    Explanations

    words related to quality, assurance, and performance

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to quality and brand assurance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iatus
    -0.83
     Pett
    -0.71
     Reported
    -0.65
     Volcano
    -0.64
     Paragu
    -0.63
     Zimbabwe
    -0.63
    amon
    -0.63
    abouts
    -0.62
    anamo
    -0.62
     Sister
    -0.61
    POSITIVE LOGITS
     ideally
    1.05
     preferably
    1.01
     strive
    1.00
     excellence
    0.91
     flavorful
    0.91
    Quality
    0.91
     attractive
    0.88
     conducive
    0.88
     achievable
    0.88
    cellence
    0.86
    Activations Density 1.064%

    No Known Activations