INDEX

Explanations

words related to interviews or statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of individuals and possibly organizations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

geries

-0.81

soDeliveryDate

-0.77

 Newsletter

-0.63

$$$$

-0.62

 Frie

-0.62

photo

-0.62

 Membership

-0.61

orship

-0.60

 Pigs

-0.60

 Wedding

-0.60

POSITIVE LOGITS

 explained

1.16

 reiterated

1.12

 stated

1.10

 elaborated

1.07

 asserted

1.07

 emphasized

1.06

 discusses

1.06

 discussed

1.04

 revealed

1.04

 outlined

1.03

Activations Density 0.287%

words related to interviews or statements made by individuals

proper nouns, particularly names of individuals and possibly organizations

No Comments

No Known Activations