INDEX

Explanations

words related to specific locations or events, particularly related to regions in the Middle East

oai_token-act-pair · gpt-3.5-turbo

mentions of locations and specific organizations or entities

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

 footprints

-0.60

ãĤ¨ãĥ«

-0.56

onz

-0.56

 Ambro

-0.54

imil

-0.54

 fate

-0.54

creen

-0.53

omething

-0.53

 fortun

-0.53

vez

-0.52

POSITIVE LOGITS

hof

0.64

agher

0.55

reciation

0.53

aceutical

0.52

inness

0.51

cussion

0.50

achine

0.48

 Hung

0.48

VG

0.47

 Cabin

0.47

Activations Density 0.932%

words related to specific locations or events, particularly related to regions in the Middle East

mentions of locations and specific organizations or entities

No Comments

No Known Activations