INDEX

Explanations

adjectives describing the nature or quality of something

oai_token-act-pair · gpt-3.5-turbo

negative evaluations or critiques of various subjects

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

UF

-0.71

aha

-0.69

OME

-0.69

¬¼

-0.69

uay

-0.69

tz

-0.68

ickets

-0.66

rero

-0.66

vation

-0.65

asio

-0.65

POSITIVE LOGITS

 albeit

1.15

but

0.96

 although

0.94

 namely

0.93

 however

0.93

 though

0.92

 except

0.91

 whereas

0.88

 insofar

0.87

huh

0.78

Activations Density 0.336%

adjectives describing the nature or quality of something

negative evaluations or critiques of various subjects

No Comments

No Known Activations