INDEX

Explanations

phrases related to laws, regulations, and official actions

oai_token-act-pair · gpt-3.5-turbo

statements about capabilities or the state of being

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 targ

-0.62

)|

-0.61

 Meridian

-0.57

 hatch

-0.57

 Swed

-0.57

 derby

-0.55

 trophy

-0.54

 retrie

-0.54

 sweats

-0.54

",

-0.54

POSITIVE LOGITS

ĸļ

0.80

yip

0.75

laun

0.72

xtap

0.72

hap

0.70

hement

0.69

ŃĶ

0.66

ARA

0.65

 nonetheless

0.64

etheless

0.63

Activations Density 0.336%

phrases related to laws, regulations, and official actions

statements about capabilities or the state of being

No Comments

No Known Activations