INDEX

Explanations

phrases related to terms and agreements

oai_token-act-pair · gpt-3.5-turbo

references to "terms" related to agreements, conditions, and policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

©¶æ¥µ

-0.76

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

-0.74

âĹ¼

-0.70

OWN

-0.68

OTS

-0.66

 Towns

-0.66

ILLE

-0.65

Assembly

-0.64

stanbul

-0.64

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.63

POSITIVE LOGITS

 terms

0.96

pring

0.87

ually

0.82

terms

0.79

mith

0.78

 Terms

0.77

poons

0.77

ional

0.77

cale

0.75

marks

0.75

Activations Density 0.014%

phrases related to terms and agreements

references to "terms" related to agreements, conditions, and policies

No Comments

No Known Activations