INDEX

Explanations

phrases or sentences introducing a topic or shifting focus in a discussion

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce new topics or questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

arium

-0.77

ãĥ´

-0.71

Alert

-0.66

Store

-0.64

KO

-0.64

ãĤ·

-0.64

aya

-0.61

harm

-0.61

 drowned

-0.60

cats

-0.60

POSITIVE LOGITS

ibaba

0.85

 Conclusion

0.84

 TABLE

0.82

ï¸

0.76

ETHOD

0.76

APTER

0.74

 secondly

0.73

FIN

0.73

 Summary

0.72

sembly

0.72

Activations Density 0.269%

phrases or sentences introducing a topic or shifting focus in a discussion

phrases that introduce new topics or questions

No Comments

No Known Activations