INDEX

Explanations

terms related to instructions or technical details

oai_token-act-pair · gpt-3.5-turbo

words related to measurements and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Brist

-0.50

 Judd

-0.50

Inv

-0.50

Zac

-0.47

 Fraz

-0.45

Hug

-0.45

 Volunte

-0.44

gew

-0.43

ĵĺ

-0.43

————————

-0.43

POSITIVE LOGITS

utenberg

0.50

livious

0.49

endif

0.48

 derog

0.47

osuke

0.46

yang

0.45

zon

0.45

lings

0.44

ongyang

0.44

ples

0.44

Activations Density 1.256%

terms related to instructions or technical details

words related to measurements and conditions

No Comments

No Known Activations