INDEX

Explanations

phrases related to direction and location in a specific context

oai_token-act-pair · gpt-3.5-turbo

directional references related to travel and roadways

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

ASED

-0.80

HAEL

-0.79

ptive

-0.79

MAT

-0.78

adden

-0.74

sonian

-0.73

ovember

-0.73

RON

-0.73

otrop

-0.72

utical

-0.70

POSITIVE LOGITS

 lanes

1.24

 Interstate

0.90

 motorists

0.89

 traffic

0.89

 lane

0.86

bound

0.83

 highways

0.81

 Amtrak

0.80

 locom

0.78

 buses

0.78

Activations Density 0.016%

phrases related to direction and location in a specific context

directional references related to travel and roadways

No Comments

No Known Activations