INDEX

Explanations

phrases or sentences where there is confusion or a mix-up in information

oai_token-act-pair · gpt-3.5-turbo

instances of confusion or misunderstanding in various contexts

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

gone

-0.70

apons

-0.70

bors

-0.69

©¶æ

-0.67

metic

-0.66

 helicop

-0.65

odder

-0.64

ternity

-0.64

eatures

-0.64

grad

-0.64

POSITIVE LOGITS

ingly

0.94

 Leilan

0.91

 Beir

0.80

ively

0.78

 confuse

0.76

INESS

0.74

ly

0.73

ilde

0.71

 Nunes

0.69

uously

0.68

Activations Density 0.024%

phrases or sentences where there is confusion or a mix-up in information

instances of confusion or misunderstanding in various contexts

No Comments

No Known Activations