INDEX

Explanations

instances of the word "dual" in the text

oai_token-act-pair · gpt-3.5-turbo

references to dual functionality or dual features in products

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orio

-0.63

################

-0.62

UFF

-0.62

FORE

-0.60

gone

-0.59

 Continued

-0.59

 Helpful

-0.58

 Warehouse

-0.58

otle

-0.58

ensen

-0.58

POSITIVE LOGITS

ity

1.05

role

0.84

 wielding

0.83

Shock

0.82

ities

0.81

 elig

0.81

vertisement

0.78

shock

0.76

ples

0.76

handedly

0.74

Activations Density 0.033%

instances of the word "dual" in the text

references to dual functionality or dual features in products

No Comments

No Known Activations