INDEX

Explanations

words related to strong negative emotions, especially anger

oai_token-act-pair · gpt-3.5-turbo

references to strong negative emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•layer 4 - anger

No Comments

Negative Logits

aneous

-0.69

raught

-0.68

ħ

-0.67

fman

-0.65

RECT

-0.65

wana

-0.64

ress

-0.63

onomy

-0.63

reads

-0.63

eering

-0.63

POSITIVE LOGITS

earchers

0.94

merga

0.85

OURCE

0.78

boro

0.74

itant

0.72

osponsors

0.71

imura

0.69

ulkan

0.69

Kes

0.69

ãĤ»

0.68

Activations Density 0.024%

words related to strong negative emotions, especially anger

references to strong negative emotions

No Comments

No Known Activations