INDEX

Explanations

words related to different locations and events, such as cities, stadiums, and courts

oai_token-act-pair · gpt-3.5-turbo

references to sports and specific events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãĥĥãĥī

-0.86

govtrack

-0.81

©¶æ¥µ

-0.73

-0.70

Ļ½

-0.69

£ı

-0.67

 untold

-0.66

etheless

-0.66

Ī

-0.63

CONCLUS

-0.61

POSITIVE LOGITS

 Photographer

0.90

 REUTERS

0.88

 Courtesy

0.83

 pose

0.81

 illustration

0.80

 gestures

0.78

 Credit

0.78

window

0.78

 circa

0.78

 Photograph

0.77

Activations Density 0.356%

words related to different locations and events, such as cities, stadiums, and courts

references to sports and specific events

No Comments

No Known Activations