INDEX

Explanations

adjectives and verbs related to sudden or unexpected events

oai_token-act-pair · gpt-3.5-turbo

terms related to economic or structural decline and the effects of precipitating factors

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

Not in Any Lists

No Comments

Negative Logits

erness

-0.73

buster

-0.68

Ĥ¬

-0.66

ership

-0.65

Reviewer

-0.65

masters

-0.64

pocket

-0.63

 burgers

-0.63

 jeans

-0.62

 Wing

-0.61

POSITIVE LOGITS

itous

1.58

itate

1.22

itating

1.21

 precip

1.19

itates

1.16

itated

1.14

ice

1.11

isions

0.98

abul

0.94

itant

0.94

Activations Density 0.020%

adjectives and verbs related to sudden or unexpected events

terms related to economic or structural decline and the effects of precipitating factors

No Comments

No Known Activations