© 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. 16873
    Prev
    Next
    INDEX
    Explanations

    proper nouns with a possession structure, particularly city or location names

    oai_token-act-pair · gpt-3.5-turbo

    possessive forms indicating significant entities 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
    Reference
    -0.79
    CLAIM
    -0.69
    Kind
    -0.68
    reference
    -0.67
    insk
    -0.67
     Canaver
    -0.66
     assume
    -0.65
    linked
    -0.65
     assumes
    -0.64
     Comparison
    -0.64
    POSITIVE LOGITS
     skyline
    1.21
     streets
    1.18
     suburbs
    1.11
     inhabitants
    1.06
     borders
    1.01
     Chinatown
    1.01
     waterways
    1.00
     neighborhoods
    1.00
     residents
    0.99
     finest
    0.99
    Activations Density 0.183%

    No Known Activations