INDEX

Explanations

conjunctions, specifically the word "but."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "but."

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

"but"

Explanation Uploaded by User

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rolls

-0.66

 Combine

-0.62

psy

-0.60

velt

-0.60

 Antiqu

-0.58

 Liberties

-0.57

vance

-0.56

wake

-0.55

 Peoples

-0.55

 Pigs

-0.54

POSITIVE LOGITS

tons

1.24

chery

1.21

chers

1.02

ler

0.95

 nonetheless

0.89

tered

0.89

 nevertheless

0.89

lers

0.88

 alas

0.84

still

0.84

Activations Density 0.138%

conjunctions, specifically the word "but."

instances of the word "but."

"but"

No Comments

No Known Activations