Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 2823
    Prev
    Next
    INDEX
    Explanations

    phrases involving negative or intense emotions

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions that connect negative sentiments or troubling events

    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
    ESE
    -0.89
    ebus
    -0.88
    ETS
    -0.86
    regation
    -0.85
    uese
    -0.78
    ederation
    -0.77
    lication
    -0.76
    ynthesis
    -0.74
    ento
    -0.74
    ocobo
    -0.74
    POSITIVE LOGITS
     awe
    1.22
     unpredictable
    1.10
     exhilar
    1.10
     frightening
    1.04
     thrilling
    1.02
     surreal
    1.02
     poignant
    1.01
     heartbreaking
    1.01
     hilarious
    1.01
     tragic
    1.00
    Activations Density 0.191%

    No Known Activations