INDEX

Explanations

unrelated characters or strings that may not be relevant to a specific pattern or topic

oai_token-act-pair · gpt-3.5-turbo

references to statistical analysis and cohort studies in research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 freel

-0.75

 Unicorn

-0.74

 Greg

-0.72

 swoop

-0.72

gre

-0.72

 hers

-0.71

 everywhere

-0.70

 neighb

-0.69

 lett

-0.69

 Healer

-0.68

POSITIVE LOGITS

Acknowled

1.40

RESULTS

1.32

Discussion

1.32

Figure

1.30

Introduction

1.30

FIG

1.28

TABLE

1.26

Conclusion

1.25

METHOD

1.24

CONCLUS

1.22

Activations Density 0.316%

unrelated characters or strings that may not be relevant to a specific pattern or topic

references to statistical analysis and cohort studies in research

No Comments

No Known Activations