© 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. 11-RES-JB
    11. 10164
    Prev
    Next
    INDEX
    Explanations

    German words, specifically related to politics and possibly diplomacy

    oai_token-act-pair · gpt-3.5-turbo

    fragmented or incomplete phrases that are indicative of uncertainty or hesitation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bending
    -0.72
     TTC
    -0.71
    ACP
    -0.70
    Phoenix
    -0.69
    independence
    -0.68
    advertisement
    -0.66
     streetcar
    -0.66
    PC
    -0.65
    Toronto
    -0.65
    Canadian
    -0.65
    POSITIVE LOGITS
     dies
    1.20
     die
    1.19
     Sie
    1.12
     aber
    1.11
     von
    1.08
    beit
    1.07
     der
    1.07
     Die
    1.04
     Das
    1.04
     Bundes
    1.03
    Activations Density 0.076%

    No Known Activations