© 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. 1-RES-JB
    11. 16454
    Prev
    Next
    INDEX
    Explanations

    it looks for words related to emotional tension and conflict

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with tension and unresolved conflict

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bus
    -0.84
    plex
    -0.84
    emouth
    -0.84
    braska
    -0.84
    visors
    -0.83
    ijah
    -0.82
    brates
    -0.81
    pillar
    -0.81
    frey
    -0.80
    brate
    -0.78
    POSITIVE LOGITS
     peril
    0.87
    BY
    0.85
     :=
    0.77
     fraught
    0.76
    LY
    0.75
     Caption
    0.74
    æĸ¹
    0.73
    orio
    0.72
     RIS
    0.67
     Canaver
    0.67
    Activations Density 0.023%

    No Known Activations