INDEX

Explanations

phrases containing the term "no-tip" or "no-tip policy"

oai_token-act-pair · gpt-3.5-turbo

negations or phrases related to "no" or "not."

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ãĥ¼ãĥĨãĤ£

-0.80

ãĥ¼ãĥĨ

-0.78

 Reloaded

-0.76

 Strait

-0.75

 Extras

-0.75

 Fork

-0.72

ãĤ¼ãĤ¦ãĤ¹

-0.71

)=(

-0.69

çĶ°

-0.69

Lod

-0.68

POSITIVE LOGITS

brainer

0.99

repeat

0.98

sized

0.97

 whatsoever

0.97

scale

0.97

issue

0.95

purpose

0.94

reply

0.93

zero

0.92

place

0.91

Activations Density 0.017%

phrases containing the term "no-tip" or "no-tip policy"

negations or phrases related to "no" or "not."

No Known Activations