© 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. 4-RES-JB
    11. 6077
    Prev
    Next
    INDEX
    Explanations

    specific locations, such as street names, addresses, and landmarks

    oai_token-act-pair · gpt-3.5-turbo

    locations and geographic features

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

    No Comments

    Negative Logits
     analysed
    -0.88
     learnt
    -0.85
     maths
    -0.80
     hitherto
    -0.80
     scept
    -0.79
     copied
    -0.79
     ana
    -0.78
    olkien
    -0.76
     epigen
    -0.76
     analys
    -0.75
    POSITIVE LOGITS
     Creek
    1.29
    town
    1.15
    ville
    1.12
     Avenue
    1.10
     Township
    1.06
     Beach
    1.04
     Harbor
    1.04
     Falls
    1.03
     Ridge
    1.02
     Boulevard
    1.02
    Activations Density 0.517%

    No Known Activations