INDEX

Explanations

verbs related to perception or understanding

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

cade

-0.68

 agric

-0.65

cious

-0.64

 whistle

-0.63

stead

-0.61

 suscept

-0.60

trap

-0.58

EStreamFrame

-0.58

 withd

-0.57

 regist

-0.56

POSITIVE LOGITS

Ĺ

0.96

dust

0.96

awed

0.91

 visions

0.88

fit

0.88

¿½

0.81

lights

0.79

 parallels

0.78

eye

0.78

enance

0.77

Activations Density 0.088%

verbs related to perception or understanding

No Comments

No Known Activations