INDEX

Explanations

phrases related to being impressed or receiving information

oai_token-act-pair · gpt-3.5-turbo

expressions of expectation, surprise, or ability related to various outcomes or experiences

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Haw

-0.72

Newsletter

-0.72

é¾įå¥ĳå£«

-0.65

 polygamy

-0.63

phis

-0.63

nery

-0.61

catentry

-0.61

Actor

-0.61

oldemort

-0.61

 Lawson

-0.61

POSITIVE LOGITS

 GREAT

0.78

favorite

0.75

wagen

0.74

Ĥ¬

0.71

ifferent

0.70

 rewarded

0.70

 noticed

0.69

VO

0.68

 SERV

0.67

itle

0.67

Activations Density 0.249%

phrases related to being impressed or receiving information

expressions of expectation, surprise, or ability related to various outcomes or experiences

No Comments

No Known Activations