INDEX

Explanations

words or prefixes related to medical or governmental entities

oai_token-act-pair · gpt-3.5-turbo

references to dioceses or hierarchical structures within religious contexts

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

oos

-0.81

witz

-0.69

eers

-0.68

suits

-0.66

suit

-0.65

munition

-0.63

ETS

-0.62

 Smoke

-0.62

tty

-0.61

 Rudd

-0.60

POSITIVE LOGITS

urnal

1.05

ocese

1.03

ablo

0.91

plom

0.90

inity

0.86

onymous

0.83

agonal

0.80

agram

0.80

ploma

0.75

abetes

0.75

Activations Density 0.035%

words or prefixes related to medical or governmental entities

references to dioceses or hierarchical structures within religious contexts

No Comments

No Known Activations