INDEX

Explanations

references to diplomats

oai_token-act-pair · gpt-3.5-turbo

mentions of diplomats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Hath

-0.70

closed

-0.68

Supplement

-0.67

omi

-0.65

forest

-0.65

efer

-0.64

THER

-0.63

YES

-0.63

thritis

-0.63

alach

-0.62

POSITIVE LOGITS

 diplomat

1.17

 diplomats

1.15

glers

1.02

 diplomacy

0.91

 envoy

0.89

 corps

0.88

 abroad

0.85

 diplomatic

0.83

 ambassador

0.83

 ambassadors

0.82

Activations Density 0.010%

references to diplomats

mentions of diplomats

No Comments

No Known Activations