INDEX

Explanations

words related to locations or places

oai_token-act-pair · gpt-3.5-turbo

references to places or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

enegger

-0.78

senal

-0.73

ueller

-0.69

aceous

-0.67

anwhile

-0.66

ENDED

-0.65

 VIDE

-0.65

 âĶľ

-0.65

obic

-0.65

 degrade

-0.63

POSITIVE LOGITS

sburg

0.94

 Borough

0.81

own

0.78

ership

0.78

ards

0.77

eus

0.75

orship

0.75

olves

0.74

shire

0.74

esses

0.73

Activations Density 0.008%

words related to locations or places

references to places or locations

No Comments

No Known Activations