INDEX

Explanations

companies, organizations, industries, and bodies mentioned in a financial or business context

oai_token-act-pair · gpt-3.5-turbo

terms related to organizations and agencies across various sectors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ultimate

-0.73

laughs

-0.68

otos

-0.67

hattan

-0.65

inges

-0.62

hands

-0.62

liest

-0.62

erous

-0.62

armac

-0.61

fortable

-0.60

POSITIVE LOGITS

are

1.12

 have

1.09

 converge

1.05

 continue

1.01

 operate

1.01

 rely

0.98

 dominate

0.97

 comprise

0.94

 collaborate

0.94

 were

0.93

Activations Density 0.349%

companies, organizations, industries, and bodies mentioned in a financial or business context

terms related to organizations and agencies across various sectors

No Comments

No Known Activations