INDEX

Explanations

mentions of military ranks and units, specifically brigades

oai_token-act-pair · gpt-3.5-turbo

references to military ranks and units

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ographed

-0.69

Tax

-0.68

 Helsinki

-0.67

eways

-0.67

 Lamar

-0.62

 Gork

-0.62

kj

-0.61

 Ceres

-0.60

eking

-0.59

onomy

-0.59

POSITIVE LOGITS

adier

1.21

itte

1.08

adian

0.98

ments

0.96

 brig

0.94

ading

0.92

inally

0.90

gae

0.89

antine

0.88

ModLoader

0.87

Activations Density 0.051%

mentions of military ranks and units, specifically brigades

references to military ranks and units

No Comments

No Known Activations