INDEX

Explanations

phrases emphasizing specific words or expressions

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to definitions and usage of words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

vertisement

-0.67

jri

-0.65

outube

-0.63

cific

-0.61

 valves

-0.61

 trave

-0.61

 millenn

-0.61

oÄŁ

-0.59

 compensated

-0.59

Attempts

-0.58

POSITIVE LOGITS

"-

1.08

\"

1.03

1.00

 acron

0.97

"_

0.96

"(

0.95

0.91

"#

0.90

``

0.88

"+

0.87

Activations Density 0.367%

phrases emphasizing specific words or expressions

terms and phrases related to definitions and usage of words

No Comments

No Known Activations