INDEX

Explanations

phrases mentioning historical contexts or using the word "historically"

oai_token-act-pair · gpt-3.5-turbo

references to historical patterns or trends

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

him

-0.85

plan

-0.76

 Sahara

-0.74

tein

-0.74

mails

-0.74

ity

-0.72

er

-0.72

gur

-0.71

ger

-0.70

ion

-0.69

POSITIVE LOGITS

 housed

0.91

orically

0.83

 conduc

0.81

 represented

0.80

 conclud

0.80

©¶æ¥µ

0.80

 speaking

0.80

 endowed

0.78

©¶æ

0.78

 oppressed

0.78

Activations Density 0.010%

phrases mentioning historical contexts or using the word "historically"

references to historical patterns or trends

No Comments

No Known Activations