INDEX

Explanations

numbers related to quantities or statistics

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

numerical expressions and their contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 virtually

-0.47

 theirs

-0.47

 especially

-0.46

ification

-0.45

 practically

-0.45

ique

-0.44

 ironically

-0.44

ivation

-0.44

 personally

-0.44

 maintaining

-0.43

POSITIVE LOGITS

 dstg

0.77

Mbps

0.75

 litres

0.74

fps

0.74

km

0.74

dB

0.73

 crore

0.73

mA

0.73

 tonnes

0.70

lbs

0.70

Activations Density 0.312%

numbers related to quantities or statistics

numerical expressions and their contexts

No Comments

No Known Activations