© 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. 8-RES-JB
    11. 20152
    Prev
    Next
    INDEX
    Explanations

    personal emotion and sentiment-related words and phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases that introduce or emphasize significant events or elements within a narrative

    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
    grounds
    -0.92
    onge
    -0.81
    encies
    -0.80
    Events
    -0.75
    dates
    -0.74
    adj
    -0.74
     existed
    -0.73
     experimented
    -0.72
    Monitor
    -0.71
    places
    -0.69
    POSITIVE LOGITS
     bang
    1.32
     vengeance
    1.17
     barrage
    1.04
     flurry
    1.04
     newfound
    0.95
     handshake
    0.92
     slew
    0.91
     flourish
    0.91
     triumphant
    0.91
     plethora
    0.89
    Activations Density 0.253%

    No Known Activations