© 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. 10-RES-JB
    11. 20269
    Prev
    Next
    INDEX
    Explanations

    phrases related to geographical locations, specifically directions

    oai_token-act-pair · gpt-3.5-turbo

    geographic directional phrases, particularly focusing on "north" and "south" references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     scissors
    -0.79
     Logo
    -0.75
     Masquerade
    -0.73
     ceremony
    -0.70
     Calculator
    -0.69
     Revolution
    -0.68
     Creator
    -0.66
     Collider
    -0.65
     Swordsman
    -0.64
     Dynamics
    -0.64
    POSITIVE LOGITS
    east
    1.79
    west
    1.65
    south
    1.50
    facing
    1.48
    western
    1.46
    central
    1.37
    East
    1.27
    north
    1.24
    heast
    1.23
    Eastern
    1.21
    Activations Density 0.027%

    No Known Activations