INDEX

Explanations

phrases indicating admiration or strong liking for something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating fandom or allegiance to various subjects

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

 accounted

-0.68

 dispatch

-0.64

opard

-0.64

COMPLE

-0.61

hole

-0.61

ItemImage

-0.59

cog

-0.58

?]

-0.58

BUS

-0.58

ural

-0.57

POSITIVE LOGITS

0.85

 sorts

0.78

irlf

0.77

 ours

0.69

mire

0.68

 liberty

0.65

etheless

0.65

 whichever

0.64

 hers

0.64

 yours

0.62

Activations Density 0.078%

phrases indicating admiration or strong liking for something

phrases indicating fandom or allegiance to various subjects

No Comments

No Known Activations