INDEX

Explanations

words related to medical emergencies or situations like "critical condition."

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

owship

-0.83

boats

-0.79

ila

-0.78

UAL

-0.76

boat

-0.75

DOM

-0.71

Pic

-0.71

 Polo

-0.70

©¶æ

-0.68

uph

-0.68

POSITIVE LOGITS

 importance

0.88

 acclaim

0.87

jun

0.84

 faculties

0.83

ity

0.83

 infrastructure

0.80

 appraisal

0.80

izes

0.80

 thinkers

0.77

 mass

0.76

Activations Density 0.028%

words related to medical emergencies or situations like "critical condition."

No Comments

No Known Activations