INDEX

Explanations

phrases related to posing questions or soliciting opinions

oai_token-act-pair · gpt-3.5-turbo

questions or discussions surrounding conditional scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

thal

-0.75

Eye

-0.72

rome

-0.71

aging

-0.71

GM

-0.71

viron

-0.69

ija

-0.69

ursed

-0.68

raught

-0.67

ults

-0.67

POSITIVE LOGITS

soever

1.26

 whether

0.74

whether

0.73

 nodd

0.72

 crooked

0.68

terday

0.64

ornia

0.63

 confir

0.62

yip

0.61

 consciously

0.60

Activations Density 0.032%

phrases related to posing questions or soliciting opinions

questions or discussions surrounding conditional scenarios

No Comments

No Known Activations