INDEX

Explanations

phrases related to actions or intentions taken by a group of people

oai_token-act-pair · gpt-3.5-turbo

phrases that denote intentions or actions by a group of people

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

pora

-0.67

aml

-0.65

isal

-0.64

ggles

-0.61

por

-0.61

éĹ

-0.61

cedented

-0.60

requent

-0.59

Growing

-0.57

è¦ļéĨĴ

-0.57

POSITIVE LOGITS

 themselves

1.41

 THEIR

0.93

 their

0.91

selves

0.88

 selves

0.85

 asses

0.75

their

0.74

 noses

0.73

Their

0.72

us

0.71

Activations Density 0.655%

phrases related to actions or intentions taken by a group of people

phrases that denote intentions or actions by a group of people

No Comments

No Known Activations