INDEX

Explanations

concluding phrases or words signaling the end of a text

oai_token-act-pair · gpt-3.5-turbo

conclusive statements or transitions in the text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 prior

-0.66

lie

-0.60

 breeding

-0.60

 dating

-0.60

 south

-0.60

 living

-0.60

 painting

-0.59

 vast

-0.58

 catering

-0.58

 stem

-0.57

POSITIVE LOGITS

 Finally

3.57

Finally

3.35

 Lastly

3.15

Lastly

2.72

 finally

1.74

 Eventually

1.66

Eventually

1.50

 Again

1.44

Conclusion

1.44

 Ultimately

1.42

Activations Density 0.019%

concluding phrases or words signaling the end of a text

conclusive statements or transitions in the text

No Comments

No Known Activations