INDEX

Explanations

phrases related to rings and circles

oai_token-act-pair · gpt-3.5-turbo

mentions of legal entities or law firms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

renheit

-0.98

ó

-0.91

ussen

-0.87

 Zucker

-0.83

Wik

-0.82

DEP

-0.80

 McCl

-0.79

Snap

-0.78

Dem

-0.77

iosyn

-0.77

POSITIVE LOGITS

 ring

2.29

 rings

2.24

Ring

2.18

 Ring

2.14

 Rings

1.98

rings

1.76

 jewelry

1.62

 diamond

1.55

ring

1.54

 necklace

1.48

Activations Density 0.189%

phrases related to rings and circles

mentions of legal entities or law firms

No Comments

No Known Activations