INDEX

Explanations

words related to strong negative emotions such as hate or loathing.

oai_token-act-pair · gpt-3.5-turbo

instances of the word "loathe" and its variations

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

Not in Any Lists

No Comments

Negative Logits

rity

-0.85

rition

-0.82

sonian

-0.80

 Norn

-0.79

ITAL

-0.76

*/(

-0.76

glass

-0.73

lished

-0.71

ãĤ¼ãĤ¦ãĤ¹

-0.71

rations

-0.71

POSITIVE LOGITS

zzle

1.02

veland

1.01

aning

0.97

aned

0.96

zz

0.94

oser

0.94

ppy

0.91

fty

0.90

aves

0.90

igh

0.88

Activations Density 0.005%

words related to strong negative emotions such as hate or loathing.

instances of the word "loathe" and its variations

No Comments

No Known Activations