INDEX

Explanations

phrases that describe a comparison or equivalence between different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate equivalences or comparisons

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

 nonetheless

-0.80

bender

-0.78

erer

-0.78

hess

-0.73

 nevertheless

-0.70

 trave

-0.67

 Became

-0.64

hran

-0.64

erers

-0.62

 furthermore

-0.62

POSITIVE LOGITS

lihood

0.94

Sov

0.74

anus

0.74

 Schr

0.64

othing

0.63

onymous

0.61

 blackmail

0.61

 subsistence

0.61

otin

0.61

 proverbial

0.58

Activations Density 0.086%

phrases that describe a comparison or equivalence between different entities or concepts

phrases that indicate equivalences or comparisons

No Comments

No Known Activations