INDEX

Explanations

contractions

oai_token-act-pair · gpt-3.5-turbo

negations or phrases emphasizing what one should avoid or not do

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

ELD

-0.79

venue

-0.75

assed

-0.73

Redd

-0.72

 srfAttach

-0.72

iday

-0.70

grounds

-0.67

liner

-0.66

former

-0.65

DIT

-0.64

POSITIVE LOGITS

 hesitate

1.19

 forget

1.15

 underestimate

1.11

 bother

1.03

 worry

1.03

 confuse

1.01

 mention

0.93

 expect

0.92

 misunderstand

0.86

 subscribe

0.85

Activations Density 0.042%

contractions

negations or phrases emphasizing what one should avoid or not do

No Comments

No Known Activations