INDEX

Explanations

dates and proper nouns related to locations

oai_token-act-pair · gpt-3.5-turbo

mentions of air force operations or military activities

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

 amplification

-0.58

 listener

-0.56

SHIP

-0.53

WT

-0.52

 Trails

-0.49

isSpecialOrderable

-0.49

trial

-0.49

 incentive

-0.49

cest

-0.48

 anonymously

-0.48

POSITIVE LOGITS

zona

0.67

pora

0.65

 womb

0.64

oola

0.63

 è£ıè

0.59

 Coffin

0.55

rays

0.55

obl

0.53

Jr

0.53

aviour

0.53

Activations Density 1.419%

dates and proper nouns related to locations

mentions of air force operations or military activities

No Comments

No Known Activations