INDEX

Explanations

words related to different levels of anger or intensity

oai_token-act-pair · gpt-3.5-turbo

the word "mad" in various contexts and expressions of emotional intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•layer 11 - anger

No Comments

Negative Logits

Lay

-0.67

Ĥİ

-0.61

çĦ

-0.60

ILA

-0.59

 Vert

-0.59

advertisement

-0.58

AFB

-0.58

 FANT

-0.57

 Citation

-0.57

 Dayton

-0.57

POSITIVE LOGITS

cap

1.24

agascar

1.17

rid

1.12

onna

1.07

der

1.02

ame

0.93

ras

0.93

rig

0.93

cow

0.90

men

0.89

Activations Density 0.026%

words related to different levels of anger or intensity

the word "mad" in various contexts and expressions of emotional intensity

No Comments

No Known Activations