© 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. 4036
    Prev
    Next
    INDEX
    Explanations

    references to locations or regions in the context of news stories

    oai_token-act-pair · gpt-3.5-turbo

    references to global perspectives or locations

    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
    ĵĺ
    -0.61
    erto
    -0.59
     Bundy
    -0.58
    ample
    -0.55
    ittees
    -0.55
    uesday
    -0.54
    resa
    -0.54
    hots
    -0.54
    hound
    -0.54
    ittee
    -0.54
    POSITIVE LOGITS
     standpoint
    1.18
     vantage
    1.03
     perspective
    1.02
     sources
    1.01
     perspectives
    0.96
     angles
    0.89
     afar
    0.88
     onwards
    0.80
     viewpoint
    0.80
     rooft
    0.77
    Activations Density 0.166%

    No Known Activations