© 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. 0-RES-JB
    11. 16516
    Prev
    Next
    INDEX
    Explanations

    words ending in '-ending' or '-ending' itself

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mole
    -0.65
    姫
    -0.64
     ox
    -0.63
     Britann
    -0.62
     LAR
    -0.61
     phot
    -0.61
     kit
    -0.60
     synerg
    -0.60
     thermal
    -0.59
     album
    -0.59
    POSITIVE LOGITS
    ending
    4.30
    ends
    2.72
    ended
    2.71
    enders
    2.59
    END
    2.22
    ender
    2.17
    end
    2.08
    ENDED
    1.73
    ension
    1.48
    ensions
    1.30
    Activations Density 0.011%

    No Known Activations