INDEX

Explanations

phrases emphasizing high intensity or strong emphasis

oai_token-act-pair · gpt-3.5-turbo

intensifiers that emphasize the degree of an attribute or quality

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

Nap

-0.65

amon

-0.61

ãĥĹ

-0.59

 Crisis

-0.59

necess

-0.59

 seiz

-0.58

Public

-0.57

 Formation

-0.57

bryce

-0.56

joy

-0.55

POSITIVE LOGITS

oppable

0.66

-|

0.64

azines

0.62

ãĤ´ãĥ³

0.60

dB

0.60

yright

0.58

efe

0.58

ossier

0.57

doors

0.57

 istg

0.57

Activations Density 0.078%

phrases emphasizing high intensity or strong emphasis

intensifiers that emphasize the degree of an attribute or quality

No Comments

No Known Activations