INDEX

Explanations

phrases related to different types of stores and shopping locations

oai_token-act-pair · gpt-3.5-turbo

locations and settings related to social activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

stood

-0.75

escription

-0.66

 plaintiffs

-0.62

ailable

-0.61

Reloaded

-0.61

STL

-0.60

 unspecified

-0.59

 Infrastructure

-0.58

 capacities

-0.57

 prevailed

-0.57

POSITIVE LOGITS

or

0.97

 cafeteria

0.79

etc

0.78

 somewhere

0.76

 cafe

0.74

 drawer

0.73

 playlist

0.72

 driveway

0.72

 cafÃ©

0.72

 checkout

0.72

Activations Density 0.551%

phrases related to different types of stores and shopping locations

locations and settings related to social activities

No Comments

No Known Activations