INDEX

Explanations

areas related to war and conflict

oai_token-act-pair · gpt-3.5-turbo

instances of expressive and emotional language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 funnel

-0.94

 payroll

-0.85

 orchestr

-0.83

 migration

-0.82

 advancing

-0.79

 aggreg

-0.79

 repatri

-0.78

 uranium

-0.78

 encomp

-0.78

 consolid

-0.78

POSITIVE LOGITS

Advertisements

1.59

Anyway

1.58

Trivia

1.55

Yeah

1.53

RAW

1.51

Reply

1.49

Anonymous

1.47

Okay

1.40

Posts

1.40

Sometimes

1.39

Activations Density 0.373%

areas related to war and conflict

instances of expressive and emotional language

No Comments

No Known Activations