INDEX

Explanations

phrases related to partnerships or collaborations

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to organizations or entities characterized by their partnerships or collaborative nature

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

TG

-0.77

press

-0.76

OG

-0.71

EMBER

-0.71

ogue

-0.68

Reilly

-0.67

alone

-0.66

ensing

-0.66

ovember

-0.64

obook

-0.64

POSITIVE LOGITS

stood

1.59

drawn

1.27

 lots

1.11

 plenty

1.10

 ample

1.08

 regard

1.07

 aspirations

1.06

 regards

1.05

standing

1.02

 respect

1.01

Activations Density 0.171%

phrases related to partnerships or collaborations

phrases that refer to organizations or entities characterized by their partnerships or collaborative nature

No Comments

No Known Activations