INDEX

Explanations

phrases related to the name "Sherrod."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "rod" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pixie

-0.67

 premiums

-0.64

 nomine

-0.63

haz

-0.62

 quarters

-0.61

anwhile

-0.60

 traged

-0.59

imates

-0.58

 citiz

-0.58

 spread

-0.57

POSITIVE LOGITS

ynamic

1.34

ynam

1.22

rod

1.17

olph

1.01

der

0.92

rome

0.90

irect

0.89

rig

0.87

ition

0.83

ollar

0.83

Activations Density 0.005%

phrases related to the name "Sherrod."

occurrences of the substring "rod" in various contexts

No Comments

No Known Activations