INDEX

Explanations

instances where the text suggests questioning or pondering about a certain topic

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that express curiosity or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

mouth

-0.76

oven

-0.70

adra

-0.69

along

-0.66

alt

-0.65

gate

-0.64

back

-0.64

ãĤ«

-0.64

together

-0.64

wash

-0.63

POSITIVE LOGITS

 finer

0.71

 suspic

0.67

 TAMADRA

0.65

 morality

0.64

 Grizz

0.64

 possible

0.64

 Juven

0.63

/+

0.62

"#

0.59

 homosexuality

0.58

Activations Density 0.149%

instances where the text suggests questioning or pondering about a certain topic

conditional phrases that express curiosity or uncertainty

No Comments

No Known Activations