INDEX

Explanations

phrases related to organizational abbreviations or acronyms like "APS"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "S"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĥ£

-0.72

wcs

-0.70

æ©Ł

-0.65

ãĤ¡

-0.63

oiler

-0.62

puter

-0.62

CPR

-0.61

 morphine

-0.61

 slee

-0.61

 stret

-0.61

POSITIVE LOGITS

ESSION

1.18

ustain

1.10

CE

1.03

ELF

1.03

OUND

1.00

IFT

0.97

UM

0.97

SE

0.97

IX

0.97

eeds

0.97

Activations Density 0.036%

phrases related to organizational abbreviations or acronyms like "APS"

occurrences of the letter "S"

No Comments

No Known Activations