INDEX

Explanations

proper nouns or phrases related to prestigious recognition or awards

oai_token-act-pair · gpt-3.5-turbo

instances of individuals being recognized or awarded

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

Cu

-0.75

bs

-0.72

EMS

-0.69

stead

-0.68

yip

-0.67

Beg

-0.65

Â·Â·

-0.62

Requ

-0.61

otropic

-0.61

 deterior

-0.60

POSITIVE LOGITS

paces

0.90

 Cosponsors

0.86

daq

0.83

essee

0.76

 finalists

0.72

 naming

0.71

*/(

0.70

ularity

0.70

lime

0.68

ames

0.68

Activations Density 0.020%

proper nouns or phrases related to prestigious recognition or awards

instances of individuals being recognized or awarded

No Comments

No Known Activations