INDEX

Explanations

word sequences related to customer feedback or complaints

oai_token-act-pair · gpt-3.5-turbo

references to feedback and interactions from the public

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Shutterstock

-0.73

 Wikimedia

-0.65

saf

-0.63

-0.62

Åį

-0.61

Ake

-0.60

utenberg

-0.59

aterasu

-0.58

ensibly

-0.58

affer

-0.58

POSITIVE LOGITS

 wondering

1.03

 testim

1.01

 wanting

0.99

 dissatisfied

0.94

 complaining

0.93

 unhappy

0.89

 stories

0.88

 complain

0.87

 frustrated

0.87

 experien

0.84

Activations Density 0.673%

word sequences related to customer feedback or complaints

references to feedback and interactions from the public

No Comments

No Known Activations