INDEX

Explanations

mentions of bombings and military attacks with explosives

oai_token-act-pair · gpt-3.5-turbo

terms related to military actions and weaponry

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

çĶ°

-0.88

âĶĢâĶĢâĶĢâĶĢ

-0.87

 Indigo

-0.86

spir

-0.85

natureconservancy

-0.80

Madison

-0.80

Redd

-0.80

Vert

-0.79

 Revel

-0.74

 Tasman

-0.74

POSITIVE LOGITS

 bombardment

1.24

 artillery

1.23

 bombard

1.20

 shelling

1.20

 weaponry

1.14

 bombed

1.13

 grenades

1.12

 bombing

1.11

 detonated

1.10

 barr

1.10

Activations Density 0.150%

mentions of bombings and military attacks with explosives

terms related to military actions and weaponry

No Comments

No Known Activations