INDEX

Explanations

words related to brands or labeling

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letters "br" 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 @ 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

 Aman

-0.64

 AMERICA

-0.64

 Relief

-0.64

FTA

-0.63

 fame

-0.63

eers

-0.63

 Modest

-0.62

meal

-0.61

 Europa

-0.60

WARE

-0.60

POSITIVE LOGITS

ackets

1.44

acket

1.38

anches

1.35

anch

1.26

igham

1.24

addock

1.17

ushing

1.13

ains

1.12

anded

1.12

ushed

1.10

Activations Density 0.013%

words related to brands or labeling

occurrences of the letters "br" in various contexts

No Comments

No Known Activations