INDEX

Explanations

the keyword "Les" in different contexts

oai_token-act-pair · gpt-3.5-turbo

references to the name "Les" and related variations in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 sailing

-0.67

INAL

-0.63

ACTED

-0.63

REL

-0.62

 sockets

-0.61

 comprehension

-0.60

 clear

-0.58

ODUCT

-0.56

 trail

-0.56

 aroused

-0.55

POSITIVE LOGITS

bians

1.28

bian

1.12

nar

1.02

lie

0.98

zek

0.95

bia

0.92

bos

0.87

ley

0.87

leys

0.86

kie

0.83

Activations Density 0.020%

the keyword "Les" in different contexts

references to the name "Les" and related variations in context

No Comments

No Known Activations